[clang] [llvm] [AMDGPU] Enable sin/cos to sincos folding for HIP (PR #181774)

[Matt Arsenault via cfe-commits]

https://github.com/arsenm requested changes to this pull request.

I'd really rather not do this. I'm working on comprehensively handling library calls. I do not want new instances of direct ocml usage anywhere 

https://github.com/llvm/llvm-project/pull/181774