Doc updates for 3.2.2

Fix Parallel Split-K on Gemm Operation Profiler (NVIDIA#1109)

* Debug and fix for parallel split-k in profiler

* restore debug files and remove prints

Add simple hash and eq methods for gemm_operations. (NVIDIA#1053)

Update README.md

Updates for 3.0 (NVIDIA#857)

Co-authored-by: Aniket Shivam <[email protected]>

New updates for 2.11 (NVIDIA#775)

* New updates.

* Minor profiler updates

Co-authored-by: Aniket Shivam <[email protected]>

CUTLASS 2.10 bug fixes and minor updates. (NVIDIA#626)

Update linear_combination_generic.h (NVIDIA#472)

add `skip_elementwise_` to support serial splitk in linear_combination_generic.h`

Update CMakeLists.txt (NVIDIA#473)

* Update CMakeLists.txt

Add 128bit int support if using nvc++ to solve NVIDIA#310 

@jeffhammond, would you please give it a try?

* Update CMakeLists.txt

correct copy paste error

Updated GEMM performance plot with CUTLASS 2.8 compiled with CUDA 11.…

…5 Toolkit (NVIDIA#375)

Updated GEMM performance plot with CUTLASS 2.8 compiled using CUDA 11.5 Toolkit.

GPUs under test:

    NVIDIA A100
    NVIDIA A2
    NVIDIA TitanV
    NVIDIA GeForce 2080 Ti