Fix SliceGradientOp to handle properly empty batches (#23784)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23784
Backward path does nothing during the gradient path when the input as empty, as
a result workspace can preserve gradient values from previous iteration and get
inconsistent inputs for some of the backward pass operators. This diff should
fix this disrepancy by always reinitializing output during the backward path.
Reviewed By: dzhulgakov
Differential Revision: D16646096
fbshipit-source-id: 8ca68dfad17a63fc87c033cce7b36b40bd77245c
