Fix/suppress warnings in kernel and kernel-laws

[satorg]

Supersedes #4187.

This PR fixes or suppresses compiler warnings in the kernel and kernel-laws modules.

[satorg]

Note: not all the warnings have been fixed. There are a bunch of warnings like these below that still remain:

1 |import scala.{specialized => sp}
  |              ^^^^^^^^^^^^^^^^^
  |              unused import

Such warnings get emitted by Scala3 only and the renamed @sp annotations are actually in use. So I believe it is a kind of a bug in Scala3, I filed the issue here: scala/scala3#20536

We can fix them by removing those import renames and using @specialized directly. But I would like to hear from the Scala3 team first.

[satorg]

I wonder if we want to not just deprecate SeqMonoid but also make it private [cats]:

cats/kernel/src/main/scala/cats/kernel/instances/SeqInstances.scala

Lines 73 to 74 in 17d7e95

	@deprecated("Use SeqMonoid.apply, which does not allocate a new instance", "2.9.0")
	class SeqMonoid[A] extends Monoid[Seq[A]] {

Looks like SeqMonoid is not supposed to be user-facing, so it should be a fairly safe change.
It would be out of scope of this PR, of course, but if it makes sense, I can do it here, because it is still related to the warnings cause.

[rossabaker]

That'd be a source breaking change for whoever used it that way. Even though they shouldn't use it, it's nice to make it very clear how to fix it. I don't tend to like to remove deprecations until keeping them causes pain.

[satorg]

I feel the best strategy here would be to provide a scalafix rule for that. But it would definitely come out of the scope of this PR.

[satorg]

Removed the [CLEANUP] prefix from the title. I feel the behind-the-scenes label is just enough for it.

[som-snytt]

I believe it is a kind of a bug

and not even the most exotic species.

Thanks for the ticket, I seized the opportunity and submitted a fix for the unused import.

I wonder if they would accept a SIP to add 2-letter aliases for all the standard annotations.

[danicheg]

lgtm

[rossabaker]

That'd be a source breaking change for whoever used it that way. Even though they shouldn't use it, it's nice to make it very clear how to fix it. I don't tend to like to remove deprecations until keeping them causes pain.