[clang] [clang][CodeGen] Emit atomic IR in place of optimized libcalls. (PR #73176)

[via cfe-commits]

Logikable wrote:

I've rewritten a part of this PR to incorporate @arichardson's suggestion. I wasn't able to find inconsistencies through a cursory review of the backend's AtomicExpand pass, nor through a small set of tests built for common targets.

How can I find the gaps I missed?

https://github.com/llvm/llvm-project/pull/73176