[llvm] [NVPTX] Generalize and extend upsizing when lowering 8/16-bit-element vector loads/stores (PR #119622)

[Artem Belevich via llvm-commits]

https://github.com/Artem-B approved this pull request.

LGTM. Thank you for the patch and I appreciate your willingness to go through all the changes during the review.

https://github.com/llvm/llvm-project/pull/119622