Cleanup opaque_type_cycle_error

[yotamofek]

Small cleanup opportunity found while reading the code: the label: bool var isn't really needed since there's only one execution path that sets is to true. Also, tried to reduce right-ward drift.

Best reviewed by hiding whitespace

[rustbot]

r? @jieyouxu

rustbot has assigned @jieyouxu.
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

[compiler-errors]

LGTM. The CI failure is unrelated.

r? compiler-errors @bors r+ rollup

[bors]

📌 Commit ad0e8dd has been approved by compiler-errors

It is now in the queue for this repository.

[bors]

🌲 The tree is currently closed for pull requests below priority 1000. This pull request will be tested once the tree is reopened.

[compiler-errors]

Gonna close/reopen to bump CI, I think this caused the rollup to fail #135317 (comment)

[compiler-errors]

@bors r-

[yotamofek]

Ahh jeez, sorry about that @compiler-errors , I was accidentally running the wrong tests locally before pushing 😅

I dived a little deeper into this function and it seems that this if is never taken:

rust/compiler/rustc_hir_analysis/src/check/check.rs

Lines 1722 to 1726 in b44e14f

	if visitor
	.returns
	.iter()
	.filter_map(|expr| typeck_results.node_type_opt(expr.hir_id))
	.all(|ty| matches!(ty.kind(), ty::Never))

i.e., I replaced its body with a panic!() and ran ./x suggest --run, and all the tests passed.
So either it's unreachable, and return paths never resolve to a ty::Never, or there's no test exercising this. So I tried to but couldn't come up with a test that does, see futile attempt here: https://godbolt.org/z/91fW8Ea5b
Any idea what's going on here?

[yotamofek]

Fixed for now, but couldn't find a decent way to not duplicate the span_label call. I'm not sure this PR is still a worthwhile improvement.