pytorch

d8506ff4 - Generalize gesvdjBatched to run whith full_matrices==false (#88502)

Generalize gesvdjBatched to run whith full_matrices==false (#88502) As brought up in https://github.com/pytorch/pytorch/issues/86234#issuecomment-1268296036, our heuristic for which SVD backend to choose was not great in some cases. The case in which there could be some improvements is when we have a large batch of very small non-square matrices. This PR, adapts the calling code to gesvdj by creating two temporary square buffers to allow to call gesvdjBatched, and then copies back the result into the output buffers. We then modify the heuristic that chooses between gesvdj and gesvdjBatched. Fixes https://github.com/pytorch/pytorch/issues/86234 Pull Request resolved: https://github.com/pytorch/pytorch/pull/88502 Approved by: https://github.com/IvanYashchuk, https://github.com/nikitaved, https://github.com/mruberry, https://github.com/xwang233