LSC 2D prefetch: Unify type of dummy data arg to i32

[silee2]

since the generated intinrsic call is independent of dummy data arg type. Type mismatch error happens if multiple callers use different types for dummy data arg.
For non 32bit element type, normalize 2D prefetch shape around i32 data type. Inner dimension gets scaled according to ratio between 32 and bitwidth of element type
Add mixed type 2D prefetch test case.

Please review these guidelines to help with the review process:

• Have you provided a meaningful PR description?
• Have you added a test, a reproducer, or a reference to an issue with a reproducer?
• Have you tested your changes locally for CPU and GPU devices?
• Have you made sure that new changes do not introduce compiler warnings?
• If this PR is a work in progress, are you filing the PR as a draft?
• Have you organized your commits logically and ensured each can be built by itself?

[mshahneo]

LGTM