Enable `ceil`, `floor`, `frac`, `round` & `trunc` for BFloat16 on CUDA (#57910)
Summary:
Enable `ceil`, `floor`, `frac`, `round` & `trunc` for BFloat16 on CUDA
Pull Request resolved: https://github.com/pytorch/pytorch/pull/57910
Reviewed By: agolynski
Differential Revision: D28393469
Pulled By: ngimel
fbshipit-source-id: b0f02ade7c6e2ed122aa5d80f6d442823dc1f221