[flake8-type-checking] Fix false positives for typing.Annotated

[Daverball]

This partially addresses #13713

Summary

Ruff properly clears the TYPE_DEFINITION flag when visiting special parts of an annotation like typing.Literal and typing.Annotated. However there are other typing related semantic flags which aren't cleared. This is fine, however is_typing_reference only check for those higher level flags and completely ignores TYPE_DEFINITION, which can lead to false positives for TCH001-003.

Test Plan

cargo nextest run

[MichaReiser]

@AlexWaygood would you mind taking a look at this PR? I think you have a better understanding of our type checking flags

[github-actions]

ruff-ecosystem results

Linter (stable)

ℹ️ ecosystem check detected linter changes. (+3 -0 violations, +0 -0 fixes in 1 projects; 53 projects unchanged)

ibis-project/ibis (+3 -0 violations, +0 -0 fixes)

+ ibis/expr/operations/generic.py:16:44: RUF100 Unused `noqa` directive (unused: `TCH001`)
+ ibis/expr/operations/generic.py:18:54: RUF100 Unused `noqa` directive (unused: `TCH001`)
+ ibis/expr/types/groupby.py:28:42: RUF100 Unused `noqa` directive (unused: `TCH001`)

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
RUF100	3	3	0	0	0

Linter (preview)

ℹ️ ecosystem check detected linter changes. (+3 -0 violations, +0 -0 fixes in 1 projects; 53 projects unchanged)

ibis-project/ibis (+3 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview

+ ibis/expr/operations/generic.py:16:44: RUF100 Unused `noqa` directive (unused: `TCH001`)
+ ibis/expr/operations/generic.py:18:54: RUF100 Unused `noqa` directive (unused: `TCH001`)
+ ibis/expr/types/groupby.py:28:42: RUF100 Unused `noqa` directive (unused: `TCH001`)

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
RUF100	3	3	0	0	0

[Daverball]

Not quite sure what the new LOG015 violations are about and how they are related to this change, but they look correct to me.

The RUF100 violations are encouraging, since these are all due to previous false positives.

[MichaReiser]

@Daverball can you try rebasing/merging main. I suspect that will help with the LOG015 ecosystem changes

[AlexWaygood]

Thank you! This LGTM.

[dhruvmanila]

@AlexWaygood I'm assuming this is a bugfix based on the PR title but feel free to update the label if you think it's a rule change.

[AlexWaygood]

@AlexWaygood I'm assuming this is a bugfix based on the PR title but feel free to update the label if you think it's a rule change.

Yes, I think that's right! Sorry, I should have added the label.