Make the suggestion message of unused_must_use scarier.

[kpreid]

The old message, “help: use let _ = ... to ignore the resulting value”, contains no indication that this is something that you might not want to do, and in fact can be read as saying that one should do this.

The new message, “help: if you wish to disregard this recommendation and immediately drop the value, use let _ = ...”, is intended to make it clear that this is a choice, and choosing it has the particular effect of immediately dropping the value.

A possible further improvement, not part of this change, might be to distinguish between types that are must_use and have a destructor (whose side-effects are likely relevant), and types that are must_use and do not have a destructor, where the timing is irrelevant and the consequence is just ignoring information that was provided. However, not every destructor is relevant to must_use; for example, Result<(), Box<dyn Error>> has a destructor and is must_use, but the destructor is only managing memory for the error value and it has no side effects relevant to the caller (unless the error in the box has side effects of its own).

Inspired by #t-lang > Lint for assigning drop guards to _

@rustbot label +A-lints +L-unused_must_use

[rustbot]

r? @jackh726

rustbot has assigned @jackh726.
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

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 76 candidates
• Random selection from 19 candidates

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.