Fix array::IntoIter::fold to use the optimized Range::fold

[scottmcm]

It was using Iterator::by_ref in the implementation, which ended up pessimizing it enough that, for example, it didn't vectorize when we tried it in the https://rust-lang.zulipchat.com/#narrow/stream/257879-project-portable-simd/topic/Reducing.20sum.20into.20wider.20types conversation.

Demonstration that the codegen test doesn't pass on the current nightly: https://rust.godbolt.org/z/Taxev5eMn

[rust-highfive]

r? @joshtriplett

(rust-highfive has picked a reviewer for you, use r? to override)

[the8472]

Huh, interesting. It previously used try_fold instead to not go through the impl<T: Iterator> Iterator for &mut T blanket impl (#87431), which was then removed in #89774 by adding #[inline] on the blanket impl which was sufficient for the vec::bench_flat_map_collect benchmark.
But yeah, this should be fine too.

[scottmcm]

Friendly one-month review reminder ping, @joshtriplett.

[the8472]

@bors r+ rollup=never

[bors]

📌 Commit e8fc7ba has been approved by the8472

[bors]

⌛ Testing commit e8fc7ba with merge 9fbbe75...

[bors]

☀️ Test successful - checks-actions
Approved by: the8472
Pushing 9fbbe75 to master...

[rust-timer]

Finished benchmarking commit (9fbbe75): comparison url.

Summary: This benchmark run did not return any relevant results.

If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf.

@rustbot label: -perf-regression