[Device] Add dynamic fetch/reduce pipelining for reduction collectives - Simple protocol (#1861)

* Support pipelining codegen and template specialization

* Support ReduceCopy pipelining for AllReduce, ReduceScatter, and Reduce (currently enabled for bfloat16)

* Remove need for FUNC_INDEX_TOTAL

* Add pipeline field to device function key construction logic

* Avoid unneeded codegen for LL/LL64 kernels

* Modify conditions and add pipeline dtypes env

* Optimize selection for both gfx942 and gfx950

* Increase pipeline bitfield width

* Use __forceinline__ for all device functions

* Realign reduceCopy with original form

* Add opt-out option to enable perf debugs

* Remove force-reduce-pipelining option from README

* Update CHANGELOG.md

---------

Co-authored-by: Jeffrey Novotny <jnovotny@amd.com>

README.md

@@ -64,7 +64,6 @@ RCCL build & installation helper script
     -t|--tests_build           Build rccl unit tests, but do not run
        --time-trace            Plot the build time of RCCL (requires `ninja-build` package installed on the system)
        --verbose               Show compile commands
-       --force-reduce-pipeline Force reduce_copy sw pipeline to be used for every reduce-based collectives and datatypes
 ```
 
 By default, RCCL builds for all GPU targets defined in `DEFAULT_GPUS` in `CMakeLists.txt`. To target specific GPU(s), and potentially reduce build time, use `--amdgpu_targets` as a `;` separated string listing GPU(s) to target.