Use SymBool for is_contiguous computation (#92229)
This changes TensorImpl to store SymBool instead of bool. However, it doesn't actually compute these quantities symbolically (outside of some top level disjunctions.) The purpose of this PR is to make it easier to diagnose performance problems in the next PR, as after this change we can switch to guardless implementations without modifying TensorImpl.h
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/92229
Approved by: https://github.com/Skylion007, https://github.com/albanD
