Report HIP occupancy-driven grid sizes in Stream-K CkProfiler output

[ozturkosu]

Proposed changes

This update introduces grid dimension tracking for Stream-K profiling in the CK Profiler. The grid dimensions, determined by hipOccupancyMaxActiveBlocksPerMultiprocessor are now explicitly tracked and stored during kernel execution. This enhancement improves debugging, profiling, and runtime analysis for Stream-K operations.

Checklist

Please put an x into the boxes that apply. You can also fill these out after creating the PR. If you're not sure, please don't hesitate to ask.

• I have run clang-format on all changed files
• Any dependent changes have been merged

Discussion

If this is a relatively large or complex change, feel free to start a discussion by explaining why you chose the solution you did and what alternatives you considered