[llvm] [AMDGPU] Use nan as the identity for atomicrmw fmax/fmin (PR #97411)

[Jay Foad via llvm-commits]

jayfoad wrote:

> > the same operation as llvm.maxnum/minnum
> 
> The definition for that is in flux at the moment.

Then can you please revert #96739 ("AMDGPU: Remove ds_fmin/ds_fmax intrinsics")? I am trying to migrate [LLPC](https://github.com/GPUOpen-Drivers/llpc) from using the intrinsics to atomicrmw but I can't do that if I don't know the semantics of atomicrmw fmax/fmin and it's liable to change under my feet.


https://github.com/llvm/llvm-project/pull/97411