use ROCm 1.9.2 fp16 capabilities in rocBLAS and MIOpen interfaces (#14994)
Summary:
* relax MIOpen if statement to allow fp16/fp32 mixed precision training now supported by ROCm 1.9.2
* use gemm_ex API of rocBLAS in ROCm 1.9.2 instead of the previous hgemm API
* with this: enable all but one half test in test_nn
While there, fix also:
* a group convolution issue w/ MIOpen pertaining to initializing MIOpen on multi-GPU systems properly we detected while working on this
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14994
Differential Revision: D13439869
Pulled By: bddppq
fbshipit-source-id: 75e4eb51a59488882e64b5eabdc30555b25be25e