lint suboptimal_flops' mul_add in const context

[Gri-ffin]

changelog: [suboptimal_flops]: lint mul_add in const context

[rustbot]

Thanks for the pull request. A reviewer will take a look after it receives 2 community reviews.

In the meantime, we would highly appreciate if you could try to review any of PRs waiting on community reviews.

[CommanderStorm]

I mean, yes your impl would work for the moment.
Could we make it work with less maintenance required? For example via Msrv::is_stable_or_const_stable?
The fn (or its siblings) look like the thing we likely want 🤔

The added nesting by is_in_const feels a bit improperly, not sure why..

View changes since this review

[Gri-ffin]

is_stable_or_const_stable needs a DefId , but we are triggering here on basic arithmetic, mul_add is the suggestion we are emitting so its not in the HIR, and AFAIK, hardcoding the version is the standard clippy pattern to check the MSRV.

The added nesting by is_in_const feels a bit improperly, not sure why..

It was either centralizing the check in mod.rs file, or sprinkling the checks in every file (which doesn't seem to me a good idea to do for only a single case).

[CommanderStorm]

Fair, I still get confused with hir vs the rest.

View changes since this review