* Reworked Unit_hipLaunchCooperativeKernel_Basic and Unit_hipLaunchCooperativeKernelMultiDevice_Basic
* Introduce reduction_factor for coop groups tests. Fix Unit_Coalesced_Group_Tiled_Partition_Sync_Positive_Basic
* Fix always false requirement by adding a cast
* Change data type to unsigned long long to align with cuda
* Change literal type to double to ensure proper type casting
* Remove formatting comments
- Introduce performance tests to project.
- Improve benchmarking utilities.
- Delete copy constructors from Timer and Benchmark classes.
- Disable Catch2's benchmarking functionalities.
- Address review comments and add progress bar/display output to the Benchmarking tool
- Add flushing of the buffer on the benchmark name display
- Introduce command line options.
- Add allocation type to string helper function.
- Add output modifier to Benchmark class.
- Fix invalid calculation of deviation
- Update performance_common.hh
- Resolve build error on Windows by adding include for reduce and accumulate
[ROCm/hip-tests commit: 5fbbdcae68]