Annotate kernels separately

[michaelmckinsey1]

Summary

• This PR is additional Caliper regions for RAJAPerf kernels that have multiple kernel launches.
• It does the following (modify list as needed):
  • Adds additional Caliper regions at the request of @pearce8
• There is currently no synchronize for these regions for asynchronous GPU kernels. But these should be profiled with CUDA/HIP events anyway, which do not need CPU synchronization to measure GPU time.
• add function types to be able to filter out kernels instead of type=function -> type=subkernel

Examples

Polybench_JACOBI_1D has one launch per rep for poly_jacobi_1D_1 and one for poly_jacobi_1D_2. So its tree will now profile each separately:

POLYBENCH_FLOYD_WARSHALL, HALO_PACKING, HALO_EXCHANGE all have variable amount of launches, so we will append _k instead of adding k regions to the tree

All kernels with only 1 launch per rep are unchanged.

[michaelmckinsey1]

made a draft because need to add different types to these regions so we can filter them with caliper

[artv3]

@michaelmckinsey1 , why not use just use RAJA's kernel naming capability here? Example...

RAJA::forall<RAJA::cuda_exec<256>>(
    range, 
    RAJA::Name("VectorAddKernel"), // <-- Kernel Name injected here
    [=] RAJA_DEVICE (int i) {
        c[i] = a[i] + b[i];
    }
);