Add [manual_rem_euclid] lint

[evantypanski]

Closes #8883

Adds a lint for checking manual use of rem_euclid(n)

changelog: Add [manual_rem_euclid] lint

[rust-highfive]

r? @Jarcho

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

[ghost]

I would rewrite this to not have the unwrap(). Something like

let val = match int_const {
    FullInt::S(s) => s.try_into(),
    FullInt::U(u) => Some(u)
}
Some((val?, other_op))

There might be a better way of doing it.

--

Edit: Just a comment. Not a review.

[evantypanski]

Agreed, believe this is in the spirit of that 75ed0c9

[Jarcho]

You can match on the constant and skip the bounds check here. i.e.

match x {
    S(x) => x.try_into().ok(),
    U(x) => Some(x),
}

[evantypanski]

Ah, realized you can't do x % 0 because zero divisor is invalid, so the equal checks for the consts makes sure the const is nonzero so some checks were completely unnecessary, thanks

[Jarcho]

rem_euclid is only const as of 1.52. You'll need to add a call to clippy_utils::in_constant with a second msrv check.

[Jarcho]

There should be a call to in_external_macro, and an msrv test for the const version with the msrv set between the two versions. Otherwise it all looks good.

[Jarcho]

Looks good now. Thank you.

@bors r+

[bors]

📌 Commit df26c3f has been approved by Jarcho

[bors]

⌛ Testing commit df26c3f with merge e17864e...

[bors]

☀️ Test successful - checks-action_dev_test, checks-action_remark_test, checks-action_test
Approved by: Jarcho
Pushing e17864e to master...