[llvm] [LV] Avoid collapsing VF to zero when a loop has no memory ops or reductions (PR #216266)

[Pengcheng Wang via llvm-commits]

================

----------------
wangpc-pp wrote:

I just copied this from `llvm/test/Transforms/LoopVectorize/AArch64/smallest-and-widest-types.ll`. For logs, I don't know if UTC is better since it is too verbose.

https://github.com/llvm/llvm-project/pull/216266