[llvm] [NVPTX] Add float to tf32 conversion intrinsic (PR #121507)

[Alex MacLean via llvm-commits]

https://github.com/AlexMaclean commented:

Thanks for the updates! baring some minor nits this looks good to me. The approaches we adopt here, for representing rounding modes etc., will probably be used in many other future intrinsic so lets make sure we set good standards now. Curious for the thoughts of @Artem-B and @andykaylor on this.

https://github.com/llvm/llvm-project/pull/121507