[cuDNN][cuDNN RNNv8 API] Fix math type behavior in cuDNN RNN (#120277)
Adds back `CUDNN_TENSOR_OP_MATH` which was erroneously dropped by #115719
CC @malfet @ptrblck
Pull Request resolved: https://github.com/pytorch/pytorch/pull/120277
Approved by: https://github.com/drisspg
