fix(vite-node): named export should overwrite export all

[hi-ogawa]

Description

• Related fix(ssr): named export should overwrite export all vitejs/vite#19534
• Closes https://discord.com/channels/804011606160703521/1362130204595720343/1362130204595720343

Porting vitejs/vite#19534 to fix vitejs/vite#19525 on Vite-node. I forgot we need to have the same thing here when the fix is in runtime but not in transform.

The bug is now always happen when ssr transform hoists each named export vitejs/vite#18983.

Please don't delete this checklist! Before submitting the PR, please make sure you do the following:

• It's really useful if your PR references an issue where it is discussed ahead of time. If the feature is substantial or introduces breaking changes without a discussion, PR might be closed.
• Ideally, include a test that fails without this PR but passes with it.
• Please, don't make changes to pnpm-lock.yaml unless you introduce a new test example.

Tests

• Run the tests with pnpm test:ci.

Documentation

• If you introduce new functionality, document it. You can run documentation with pnpm run docs command.

Changesets

• Changes in changelog are generated from PR name. Please, make sure that it explains your changes in an understandable manner. Please, prefix changeset messages with feat:, fix:, perf:, docs:, or chore:.

[ghiscoding]

@hi-ogawa what if this PR and the indirect changes still doesn't fix the issue brought by the hoisting Vite change? Because for me in Lerna-Lite project, I still see it failing when I enable pnpm lock file maintenance and it uses all latest versions of Vite/Vitest. For now I can stay on previous Vite 6.2.x to avoid the failures (I only use Vitest in Lerna-Lite, there's no use of Vite since it's not a web app but rather a Node app)

My failure are in this GitHub Actions workflow
https://github.com/lerna-lite/lerna-lite/actions/runs/14698826781/job/41244615531

and the PR (that I now closed to avoid failures in the project)
lerna-lite/lerna-lite#1024

I get these failures only when using Vite 6.3.x, while previous 6.2.x is all good. I haven't had a chance to really dig into the problem, I just decided to stay on previous Vite for now. Should I open a new Vitest issue?

⎯⎯⎯⎯⎯⎯ Failed Suites 6 ⎯⎯⎯⎯⎯⎯⎯

 FAIL  packages/publish/src/__tests__/publish-from-git.spec.ts [ packages/publish/src/__tests__/publish-from-git.spec.ts ]
Error: [vitest] There was an error when mocking a module. If you are using "vi.mock" factory, make sure there are no top level variables inside, since this call is hoisted to top of the file. Read more: https://vitest.dev/api/vi.html#vi-mock
 ❯ packages/publish/src/__tests__/publish-from-git.spec.ts:35:1
     33| import { npmPublish } from '../lib/npm-publish.js';
     34| import { npmPublish as npmPublishMock } from '../lib/__mocks__/npm-pub…
     35| import { logOutput, promptConfirmation, PublishCommandOption, throwIfU…
       | ^
     36| 
     37| // helpers

Caused by: Error: [vitest] There was an error when mocking a module. If you are using "vi.mock" factory, make sure there are no top level variables inside, since this call is hoisted to top of the file. Read more: https://vitest.dev/api/vi.html#vi-mock
 ❯ packages/core/src/filter-packages/lib/filter-packages.ts:2:1

Caused by: ReferenceError: Cannot access 'Command' before initialization
 ❯ Module.get [as Command] packages/core/src/command.ts:1:103
 ❯ packages/publish/src/__tests__/publish-from-git.spec.ts:14:70
 ❯ packages/core/src/filter-packages/lib/filter-packages.ts:2:1

[hi-ogawa]

@ghiscoding Please open a new issue. 🙏

[EvHaus]

I'm having the same issue as @ghiscoding. Some of my tests are unexpectedly failing with these errors:

ReferenceError: Cannot access '***' before initialization

Only after upgrading from Vite 6.2.6 to 6.3.3.

[ghiscoding]

issue #7902 created, @EvHaus you can add any missing info in that issue. Thanks