feat(rwlock_atomic): add new lints for using RwLock with atomic types

[Tunglies]

changelog: [rwlock_atomic]: new lint
changelog: [rwlock_integer]: new lint

[rustbot]

r? @blyxyas

rustbot has assigned @blyxyas.
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

[Tunglies]

Hey @blyxyas, would you be interested in reviewing?

[blyxyas]

Yes @Tunglies I'll review this one, could you give us some more information in the pull request's description? Is this fixing any existing issues? In general, more context on why this pull request is needed would be very appreciated.

Thanks!

[Tunglies]

Yes @Tunglies I'll review this one, could you give us some more information in the pull request's description? Is this fixing any existing issues? In general, more context on why this pull request is needed would be very appreciated.

Thanks!

I found Mutex rulesclippy::mutex_atomic and clippy::mutex_integer rules, use them for my comunnity's project clash-verge-rev to helps reduce Arc and lock performance cost, most are Arc<Mutex/Rwlock<bool>>. I tried write some basic branchmark for Arc<Mutex/RwLock<T>> and Atomic<T> with std lib, found that Arc<RwLock<T>> is even slower than Arc<Mutex<T>>. While continue performance development, if has Rwlock rule for this usage same as clippy::mutex_atomic and clippy::mutex_integer would be more convenience.

[blyxyas]

Seems that the lint logic is really similar to mutex_atomic, could you merge these?

View changes since this review

[blyxyas]

I don't understand this paragraph. Is RwLock supposed to be Atomic here? Could you clarify? ミ๏ｖ๏彡

[blyxyas]

I think that this lints should be suspicious, what do you think? (=ↀωↀ=)

[blyxyas]

Seems that these two lints are very very similar to mutex_atomic.rs, could you merge these 4 lints? They same like 90% of the same logic

[rustbot]

Reminder, once the PR becomes ready for a review, use @rustbot ready.