Auto merge of #105409 - compiler-errors:closure-infer-cycle, r=jackh726
Don't deduce a signature that makes a closure cyclic
Sometimes when elaborating supertrait bounds for closure signature inference, we end up deducing a closure signature that is cyclical because either a parameter or the return type references a projection mentioning `Self` that also has escaping bound vars, which means that it's not eagerly replaced with an inference variable.
Interestingly, this is not *just* related to my PR that elaborates supertrait bounds for closure signature deduction. The committed test `supertrait-hint-cycle-3.rs` shows **stable** code that is fixed by this PR:
```rust
trait Foo<'a> {
type Input;
}
impl<F: Fn(u32)> Foo<'_> for F {
type Input = u32;
}