Stabilise redirected-noqa (RUF101)

[AlexWaygood]

Summary

Stabilise two one ruff rule for the 0.6 release:

• missing-f-string-syntax (RUF027)
• redirected-noqa (RUF101)

These have both been in preview for a while and there are no open issues about either of them. They also both seem like very useful rules.

I'm fixing a tiny oversight in the RUF027 logic as part of this PR, but it's never been reported and it's unlikely to come up.

Test Plan

cargo test -p ruff_linter --lib + ecosystem

[github-actions]

ruff-ecosystem results

Linter (stable)

ℹ️ ecosystem check detected linter changes. (+3 -0 violations, +0 -0 fixes in 3 projects; 51 projects unchanged)

DisnakeDev/disnake (+1 -0 violations, +0 -0 fixes)

+ disnake/utils.py:1161:56: RUF101 [*] `PGH001` is a redirect to `S307`

RasaHQ/rasa (+1 -0 violations, +0 -0 fixes)

+ tests/graph_components/validators/test_default_recipe_validator.py:815:72: RUF101 [*] `RUF011` is a redirect to `B035`

wntrblm/nox (+1 -0 violations, +0 -0 fixes)

+ nox/manifest.py:357:50: RUF101 [*] `PGH001` is a redirect to `S307`

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
RUF101	3	3	0	0	0

Linter (preview)

✅ ecosystem check detected no linter changes.

[AlexWaygood]

Most of the RUF027 hits seem to be true positives that I think projects would very much like to know about (so this shows the value of stabilising the rule)... however, this one is a false positive, that I think wouldn't be too hard for us to fix: https://github.com/pytest-dev/pytest/blob/38ad84bafd18d15ceff1960d636c693560337844/src/_pytest/main.py#L1055-L1060

[AlexWaygood]

Okay, this is another false positive: https://github.com/pypa/cibuildwheel/blob/c6dd39bd834ed458bb627c294854832e21971f1a/cibuildwheel/macos.py#L742. The rule should exclude logging calls in the same way that it excludes gettext() calls. In that case, I don't think this rule can be stabilised right now, which is a shame :(

[AlexWaygood]

I've taken the RUF027 stabilisation out of this PR. I'll put up a separate PR to reduce the false positives after the release is out.