defined compile time macro for warp size with ROCm 7

[sidamd]

Motivation

Defined compile time macro for warp size with ROCm 7.

Technical Details

Added the macro in ops.hip and kernel.hip where warpSize is being defined. Without the fix, warpSize won't be defined.
Reference:

https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_cpp_language_extensions.html#warpsize

Test Plan

Build the project with new changes on both ROCm 6.4.3 and ROCm 7.0.0. Run the pytest suite on both versions.

Test Result

pytests passed on both ROCm versions with changes added.

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.