[Performance] LSTM/GRU scan: canonical strides + cuDNN flat-storage clones + thread-local recurrent mode#3754

vmoens merged 4 commits into

gh/vmoens/270/basepytorch/rl:gh/vmoens/270/basefrom

gh/vmoens/270/headpytorch/rl:gh/vmoens/270/headCopy head branch name to clipboard