Fix false positive for unused_unit

[benschulz]

Given a type alias as follows, clippy would detect a false positive for unused_unit.

type UnusedParensButNoUnit = Box<dyn (Fn())>;

changelog: [unused_unit]: fix false positive for Fn bounds

[rustbot]

r? @llogiq

rustbot has assigned @llogiq.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[benschulz]

Perhaps this should have been covered by !hir_ty.span.from_expansion(). Certainly, there are no explicit -> () tokens in the source. If this is the case, I'm happy to close this PR and file an issue upstream.

[llogiq]

Those expansion checks are quite finicky; sometimes you have an expression without expansion but a subexpression expanded, etc. So as long as this appears to fix the issue, it's better than nothing.

[samueltardieu]

Since this is a targeted lint, could you please use #[expect(unused_parens)] instead? That doubles as an extra free test for unused_parens that would force us to think about it if its behavior changes.

[benschulz]

This does not actually emit a warning by rustc yet. I'm currently looking into why that is exactly. I can remove the attribute and change the name of the type alias if you want.

[llogiq]

Thank you.