Remove cargo test targets, use nextest exclusively (sigp#7874)

Fixes sigp#7835


  - Remove cargo test-based Make targets (`test-release`, `test-debug`, `run-ef-tests`)
- Update aliases (`test`, `test-full`, `test-ef`) to use existing nextest equivalents
- Update contributing documentation to use nextest examples
- Fix example commands that previously referenced non-existing packages (`ssz`/`eth2_ssz`)


Co-Authored-By: Daniel Ramirez-Chiquillo <[email protected]>

Author: danielrachi1

.github/workflows/test-suite.yml

@@ -102,7 +102,7 @@ jobs:
       with:
         version: nightly-ca67d15f4abd46394b324c50e21e66f306a1162d
     - name: Run tests in release
-      run: make nextest-release
+      run: make test-release
     - name: Show cache stats
       if: env.SELF_HOSTED_RUNNERS == 'true'
       continue-on-error: true
@@ -134,7 +134,7 @@ jobs:
     - name: Set LIBCLANG_PATH
       run: echo "LIBCLANG_PATH=$((gcm clang).source -replace "clang.exe")" >> $env:GITHUB_ENV
     - name: Run tests in release
-      run: make nextest-release
+      run: make test-release
     - name: Show cache stats
       if: env.SELF_HOSTED_RUNNERS == 'true'
       continue-on-error: true
@@ -269,7 +269,7 @@ jobs:
       with:
         version: nightly-ca67d15f4abd46394b324c50e21e66f306a1162d
     - name: Run tests in debug
-      run: make nextest-debug
+      run: make test-debug
     - name: Show cache stats
       if: env.SELF_HOSTED_RUNNERS == 'true'
       continue-on-error: true
@@ -306,7 +306,7 @@ jobs:
           cache-target: release
           bins: cargo-nextest
     - name: Run consensus-spec-tests with blst and fake_crypto
-      run: make nextest-ef
+      run: make test-ef
     - name: Show cache stats
       if: env.SELF_HOSTED_RUNNERS == 'true'
       continue-on-error: true

CLAUDE.md

@@ -7,34 +7,38 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 **Important**: Always branch from `unstable` and target `unstable` when creating pull requests.
 
 ### Building and Installation
+
 - `make install` - Build and install the main Lighthouse binary in release mode
 - `make install-lcli` - Build and install the `lcli` utility binary
 - `cargo build --release` - Standard Rust release build
 - `cargo build --bin lighthouse --features "gnosis,slasher-lmdb"` - Build with specific features
 
 ### Testing
+
 - `make test` - Run the full test suite in release mode (excludes EF tests, beacon_chain, slasher, network, http_api)
-- `make nextest-release` - Run tests using nextest (faster parallel test runner)
+- `make test-release` - Run tests using nextest (faster parallel test runner)
 - `make test-beacon-chain` - Run beacon chain tests for all supported forks
 - `make test-slasher` - Run slasher tests with all database backend combinations
 - `make test-ef` - Download and run Ethereum Foundation test vectors
 - `make test-full` - Complete test suite including linting, EF tests, and execution engine tests
-- `cargo test -p <package_name>` - Run tests for a specific package
-- `cargo test -p <package_name> <test_name>` - Run individual test (preferred during development iteration)
+- `cargo nextest run -p <package_name>` - Run tests for a specific package
+- `cargo nextest run -p <package_name> <test_name>` - Run individual test (preferred during development iteration)
 - `FORK_NAME=electra cargo nextest run -p beacon_chain` - Run tests for specific fork
 
 **Note**: Full test suite takes ~20 minutes. When iterating, prefer running individual tests.
 
-### Linting and Code Quality  
+### Linting and Code Quality
+
 - `make lint` - Run Clippy linter with project-specific rules
 - `make lint-full` - Run comprehensive linting including tests (recommended for thorough checking)
 - `make cargo-fmt` - Check code formatting with rustfmt
 - `make check-benches` - Typecheck benchmark code
 - `make audit` - Run security audit on dependencies
 
 ### Cross-compilation
+
 - `make build-x86_64` - Cross-compile for x86_64 Linux
-- `make build-aarch64` - Cross-compile for ARM64 Linux  
+- `make build-aarch64` - Cross-compile for ARM64 Linux
 - `make build-riscv64` - Cross-compile for RISC-V 64-bit Linux
 
 ## Architecture Overview
@@ -44,13 +48,15 @@ Lighthouse is a modular Ethereum consensus client with two main components:
 ### Core Components
 
 **Beacon Node** (`beacon_node/`)
+
 - Main consensus client that syncs with the Ethereum network
 - Contains the beacon chain state transition logic (`beacon_node/beacon_chain/`)
 - Handles networking, storage, and P2P communication
 - Provides HTTP API for validator clients and external tools
 - Entry point: `beacon_node/src/lib.rs`
 
-**Validator Client** (`validator_client/`) 
+**Validator Client** (`validator_client/`)
+
 - Manages validator keystores and performs validator duties
 - Connects to beacon nodes via HTTP API
 - Handles block proposals, attestations, and sync committee duties
@@ -60,31 +66,37 @@ Lighthouse is a modular Ethereum consensus client with two main components:
 ### Key Subsystems
 
 **Consensus Types** (`consensus/types/`)
+
 - Core Ethereum consensus data structures (BeaconState, BeaconBlock, etc.)
 - Ethereum specification implementations for different networks (mainnet, gnosis)
 - SSZ encoding/decoding and state transition primitives
 
 **Storage** (`beacon_node/store/`)
+
 - Hot/cold database architecture for efficient beacon chain storage
 - Supports multiple backends (LevelDB, RocksDB, REDB)
 - Handles state pruning and historical data management
 
 **Networking** (`beacon_node/lighthouse_network/`, `beacon_node/network/`)
+
 - Libp2p-based P2P networking stack
 - Gossipsub for message propagation
 - Discovery v5 for peer discovery
 - Request/response protocols for sync
 
 **Fork Choice** (`consensus/fork_choice/`, `consensus/proto_array/`)
+
 - Implements Ethereum's fork choice algorithm (proto-array)
 - Manages chain reorganizations and finality
 
 **Execution Layer Integration** (`beacon_node/execution_layer/`)
+
 - Interfaces with execution clients
 - Retrieves payloads from local execution layer or external block builders
 - Handles payload validation and builder integration
 
 **Slasher** (`slasher/`)
+
 - Optional slashing detection service
 - Supports LMDB, MDBX, and REDB database backends
 - Can be enabled with `--slasher` flag
@@ -120,67 +132,79 @@ Lighthouse is a modular Ethereum consensus client with two main components:
 ## Common Review Standards
 
 ### CI/Testing Requirements
+
 - All checks must pass before merge
 - Test coverage expected for significant changes
 - Flaky tests are actively addressed and fixed
 - New features often require corresponding tests
 - `beacon_chain` and `http_api` tests support fork-specific testing using `FORK_NAME` env var when `beacon_chain/fork_from_env` feature is enabled
 
 ### Code Quality Standards
+
 - Clippy warnings must be fixed promptly (multiple PRs show this pattern)
 - Code formatting with `cargo fmt` enforced
 - Must run `cargo sort` when adding dependencies - dependency order is enforced on CI
 - Performance considerations for hot paths
 
 ### Documentation and Context
+
 - PRs require clear descriptions of what and why
 - Breaking changes need migration documentation
 - API changes require documentation updates
 - When CLI is updated, run `make cli-local` to generate updated help text in lighthouse book
 - Comments appreciated for complex logic
 
 ### Security and Safety
+
 - Careful review of consensus-critical code paths
 - Error handling patterns must be comprehensive
 - Input validation for external data
 
 ## Development Patterns and Best Practices
 
 ### Panics and Error Handling
+
 - **Panics should be avoided at all costs**
 - Always prefer returning a `Result` or `Option` over causing a panic (e.g., prefer `array.get(1)?` over `array[1]`)
 - Avoid `expect` or `unwrap` at runtime - only acceptable during startup when validating CLI flags or configurations
 - If you must make assumptions about panics, use `.expect("Helpful message")` instead of `.unwrap()` and provide detailed reasoning in nearby comments
 - Use proper error handling with `Result` types and graceful error propagation
 
 ### Rayon Usage
+
 - Avoid using the rayon global thread pool as it results in CPU oversubscription when the beacon processor has fully allocated all CPUs to workers
 - Use scoped rayon pools started by beacon processor for computational intensive tasks
 
 ### Locks
+
 - Take great care to avoid deadlocks when working with fork choice locks - seek detailed review ([reference](beacon_node/beacon_chain/src/canonical_head.rs:9))
 - Keep lock scopes as narrow as possible to avoid blocking fast-responding functions like the networking stack
 
 ### Async Patterns
+
 - Avoid blocking computations in async tasks
 - Spawn a blocking task instead for CPU-intensive work
 
 ### Tracing
+
 - Design spans carefully and avoid overuse of spans just to add context data to events
 - Avoid using spans on simple getter methods as it can result in performance overhead
 - Be cautious of span explosion with recursive functions
 - Use spans per meaningful step or computationally critical step
 - Avoid using `span.enter()` or `span.entered()` in async tasks
 
 ### Database
+
 - Maintain schema continuity on `unstable` branch
 - Database migrations must be backward compatible
 
 ### Consensus Crate
+
 - Use safe math methods like `saturating_xxx` or `checked_xxx`
 - Critical that this crate behaves deterministically and MUST not have undefined behavior
 
 ### Testing Patterns
+
 - **Use appropriate test types for the right scenarios**:
   - **Unit tests** for single component edge cases and isolated logic
   - **Integration tests** using [`BeaconChainHarness`](beacon_node/beacon_chain/src/test_utils.rs:668) for end-to-end workflows
@@ -204,14 +228,17 @@ Lighthouse is a modular Ethereum consensus client with two main components:
   - See [`scripts/local_testnet/README.md`](scripts/local_testnet/README.md) for setup instructions
 
 ### TODOs and Comments
+
 - All `TODO` statements must be accompanied by a GitHub issue link
 - Prefer line (`//`) comments to block comments (`/* ... */`)
 - Use doc comments (`///`) before attributes for public items
 - Keep documentation concise and clear - avoid verbose explanations
 - Provide examples in doc comments for public APIs when helpful
 
 ## Logging Guidelines
+
 Use appropriate log levels for different scenarios:
+
 - **`crit`**: Critical issues with major impact to Lighthouse functionality - Lighthouse may not function correctly without resolving. Needs immediate attention.
 - **`error`**: Error cases that may have moderate impact to Lighthouse functionality. Expect to receive reports from users for this level.
 - **`warn`**: Unexpected code paths that don't have major impact - fully recoverable. Expect user reports if excessive warning logs occur.
@@ -221,6 +248,7 @@ Use appropriate log levels for different scenarios:
 ## Code Examples
 
 ### Safe Math in Consensus Crate
+
 ```rust
 // ❌ Avoid - could panic
 let result = a + b;
@@ -234,6 +262,7 @@ let result = a.safe_add(b)?;
 ```
 
 ### Panics and Error Handling
+
 ```rust
 // ❌ Avoid - could panic at runtime
 let value = some_result.unwrap();
@@ -253,6 +282,7 @@ let item = array.get(1).expect("Array always has at least 2 elements due to vali
 ```
 
 ### TODO Format
+
 ```rust
 pub fn my_function(&mut self, _something: &[u8]) -> Result<String, Error> {
     // TODO: Implement proper validation here
@@ -261,6 +291,7 @@ pub fn my_function(&mut self, _something: &[u8]) -> Result<String, Error> {
 ```
 
 ### Async Task Spawning for Blocking Work
+
 ```rust
 // ❌ Avoid - blocking in async context
 async fn some_handler() {
@@ -276,6 +307,7 @@ async fn some_handler() {
 ```
 
 ### Tracing Span Usage
+
 ```rust
 // ❌ Avoid - span on simple getter
 #[instrument]
@@ -291,9 +323,10 @@ async fn process_block(&self, block: Block) -> Result<(), Error> {
 ```
 
 ## Build and Development Notes
-- Full builds and tests take 5+ minutes - use large timeouts (300s+) for any `cargo build`, `cargo test`, or `make` commands
+
+- Full builds and tests take 5+ minutes - use large timeouts (300s+) for any `cargo build`, `cargo nextest`, or `make` commands
 - Use `cargo check` for faster iteration during development and always run after code changes
+- Prefer targeted package tests (`cargo nextest run -p <package>`) and individual tests over full test suite when debugging specific issues
 - Use `cargo fmt --all && make lint-fix` to format code and fix linting issues once a task is complete
-- Prefer targeted package tests (`cargo test -p <package>`) and individual tests over full test suite when debugging specific issues
 - Always understand the broader codebase patterns before making changes
-- Minimum Supported Rust Version (MSRV) is documented in `lighthouse/Cargo.toml` - ensure Rust version meets or exceeds this requirement
+- Minimum Supported Rust Version (MSRV) is documented in `lighthouse/Cargo.toml` - ensure Rust version meets or exceeds this requirement

Makefile

@@ -139,29 +139,18 @@ build-release-tarballs:
 	$(call tarball_release_binary,$(BUILD_PATH_RISCV64),$(RISCV64_TAG),"")
 
 
+
 # Runs the full workspace tests in **release**, without downloading any additional
 # test vectors.
 test-release:
-	cargo test --workspace --release --features "$(TEST_FEATURES)" \
- 		--exclude ef_tests --exclude beacon_chain --exclude slasher --exclude network \
-		--exclude http_api
-
-# Runs the full workspace tests in **release**, without downloading any additional
-# test vectors, using nextest.
-nextest-release:
 	cargo nextest run --workspace --release --features "$(TEST_FEATURES)" \
 		--exclude ef_tests --exclude beacon_chain --exclude slasher --exclude network \
 		--exclude http_api
 
+
 # Runs the full workspace tests in **debug**, without downloading any additional test
 # vectors.
 test-debug:
-	cargo test --workspace --features "$(TEST_FEATURES)" \
-		--exclude ef_tests --exclude beacon_chain --exclude network --exclude http_api
-
-# Runs the full workspace tests in **debug**, without downloading any additional test
-# vectors, using nextest.
-nextest-debug:
 	cargo nextest run --workspace --features "$(TEST_FEATURES)" \
 		--exclude ef_tests --exclude beacon_chain --exclude network --exclude http_api
 
@@ -173,15 +162,9 @@ cargo-fmt:
 check-benches:
 	cargo check --workspace --benches --features "$(TEST_FEATURES)"
 
-# Runs only the ef-test vectors.
-run-ef-tests:
-	rm -rf $(EF_TESTS)/.accessed_file_log.txt
-	cargo test --release -p ef_tests --features "ef_tests,$(EF_TEST_FEATURES)"
-	cargo test --release -p ef_tests --features "ef_tests,$(EF_TEST_FEATURES),fake_crypto"
-	./$(EF_TESTS)/check_all_files_accessed.py $(EF_TESTS)/.accessed_file_log.txt $(EF_TESTS)/consensus-spec-tests
 
-# Runs EF test vectors with nextest
-nextest-run-ef-tests:
+# Runs EF test vectors
+run-ef-tests:
 	rm -rf $(EF_TESTS)/.accessed_file_log.txt
 	cargo nextest run --release -p ef_tests --features "ef_tests,$(EF_TEST_FEATURES)"
 	cargo nextest run --release -p ef_tests --features "ef_tests,$(EF_TEST_FEATURES),fake_crypto"
@@ -233,9 +216,6 @@ test-ef: make-ef-tests run-ef-tests
 # Downloads and runs the nightly EF test vectors.
 test-ef-nightly: make-ef-tests-nightly run-ef-tests
 
-# Downloads and runs the EF test vectors with nextest.
-nextest-ef: make-ef-tests nextest-run-ef-tests
-
 # Runs tests checking interop between Lighthouse and execution clients.
 test-exec-engine:
 	make -C $(EXECUTION_ENGINE_INTEGRATION) test