[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>

[ROCm/rccl commit: 277747c199]

projects/rccl/src/include/rccl_common.h

@@ -83,6 +83,7 @@ inline size_t rcclGetSizePerRank(ncclFunc_t const& func, size_t const& nBytes, i
 }
 void rcclUpdateCollectiveProtocol(struct ncclComm* comm, size_t const& nBytes, struct ncclTaskColl* info);
 void rcclUpdateThreadThreshold(struct ncclComm* comm, size_t const& nBytes, struct ncclTaskColl* info, int& threadThreshold);
+void rcclSetPipelining(struct ncclComm* comm, size_t const& nBytes, struct ncclTaskColl* info);
 ncclResult_t rcclGetAlgoInfo(struct ncclComm* comm, ncclFunc_t coll, uint64_t count, ncclDataType_t dataType,
                              int collNetSupport, int nvlsSupport, int numPipeOps,
                              int* algo, int* protocol, int* maxChannels);