fix: Use drain() instead of keys().next() + swap_remove() in auto_tra…

[strmfos]

Replace two-step operation to get the first map entry (keys().next() + swap_remove()) with a more idiomatic and efficient single drain().next() call

[rustbot]

r? @notriddle

rustbot has assigned @notriddle.
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

[rust-log-analyzer]

The job pr-check-2 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

[RUSTC-TIMING] askama test:false 0.174
error[E0061]: this method takes 1 argument but 0 arguments were supplied
   --> src/librustdoc/clean/auto_trait.rs:283:34
    |
283 |         let (target, deps) = map.drain().next().unwrap();
    |                                  ^^^^^-- argument #1 is missing
    |
note: method defined here
   --> /cargo/registry/src/index.crates.io-1949cf8c6b5b557f/indexmap-2.10.0/src/map.rs:300:12
    |
300 |     pub fn drain<R>(&mut self, range: R) -> Drain<'_, K, V>
    |            ^^^^^
help: provide the argument
    |
283 |         let (target, deps) = map.drain(/* range */).next().unwrap();
    |                                        +++++++++++

For more information about this error, try `rustc --explain E0061`.
[RUSTC-TIMING] rustdoc test:false 3.330
error: could not compile `rustdoc` (lib) due to 1 previous error

[bjorn3]

Using drain will clear out the entire Vec. As per the docs:

If the iterator is dropped before being fully consumed, it drops the remaining removed elements.

[Dylan-DPC]

@strmfos any updates on this? thanks

[fmease]

Since the current change is incorrect, I'll close this one. I've been meaning to fully rewrite auto trait impl synthesis for ages now (and with the next solver coming closer to the finish line, I need to act very soon) which would likely make this obsolete anyway.