[red-knot] Re-enable linter corpus tests

[sharkdp]

Summary

Given that no new failures appeared over the past two weeks, and seeing the fuzzing results from @dhruvmanila in #13778, I think we can re-enable these tests. We also had one regression that would have been caught by these tests, so there is some value in having them enabled.

[carljm]

Looks fine to me in principle. Appears there is at least one file that needs adding to KNOWN_FAILURES before this can land?

[github-actions]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

✅ ecosystem check detected no linter changes.

[sharkdp]

Given that no new failures appeared over the past two weeks,

Appears there is at least one file that needs adding to KNOWN_FAILURES before this can land

Fantastic, unfortunate timing on my end 😄. I accidentally ran cargo nextest run -p red_knot_python_semantic offline, instead of -p red_knot_workspace.

This new file also contains a self-referential type alias, so this is a known problem (#14672). I added it to the corresponding section in KNOWN_FAILURES.

type K = 'K | None'