PIE794: Detect duplicated declared class fields

[MichaReiser]

Summary

Fixes #22683

The rule skipped annotated assignments without a value. This seems an unintentional change that was introduced in #8634. This PR re-enables detecting duplicated class fields where a later declaration is an annotated assignment without a value.

We might have to preview gate this change if there are many ecosystem hits. Although I doubt that, given that this is unlikely very common.

Test Plan

Added regression test.

[astral-sh-bot]

ruff-ecosystem results

Linter (stable)

ℹ️ ecosystem check detected linter changes. (+2 -0 violations, +0 -0 fixes in 1 projects; 54 projects unchanged)

reflex-dev/reflex (+2 -0 violations, +0 -0 fixes)

+ reflex/components/recharts/cartesian.py:245:5: PIE794 Class field `fill` is defined multiple times
+ reflex/components/recharts/cartesian.py:248:5: PIE794 Class field `stroke` is defined multiple times

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
PIE794	2	2	0	0	0

Linter (preview)

ℹ️ ecosystem check detected linter changes. (+2 -0 violations, +0 -0 fixes in 1 projects; 54 projects unchanged)

reflex-dev/reflex (+2 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

+ reflex/components/recharts/cartesian.py:245:5: PIE794 Class field `fill` is defined multiple times
+ reflex/components/recharts/cartesian.py:248:5: PIE794 Class field `stroke` is defined multiple times

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
PIE794	2	2	0	0	0

[MichaReiser]

@AlexWaygood I always get confused about the trillion different ways on how you can define field and class level attributes in Python. Does the usage in the reflex project look legit to you (does it define a class and field level fill attribute?)

[AlexWaygood]

The reflex hits look like true positives to me! They're basically doing this:

class Foo:
    x: int = 0
    # 10 lines later...
    x: int

and there's no reason to do that instead of just:

class Foo:
    x: int = 0

We might want to be careful about redeclarations to a different type, though? ty supports these, although no other type checker does (at least, not by default):

class Foo:
    x: int = 0
    x: str = "foo"

[MichaReiser]

We might want to be careful about redeclarations to a different type, though? ty supports these, although no other type checker does (at least, not by default):

I thought about that too but this seems like a separate change to the rule