Enable fp8 support (#1101)

* initial checkin

* resolve cr comments

* resolve the build issue

* fix the data correctless issue

* update fp8 header file and update the unit test for fp8 support

* remove fp16 from fp8 headers

* fix ut issue and catch up the latest code from develop

* udate according to cr comments

* update ut according to cr comments

* update num floats for each SumPostDiv from 4 to 6

* update fp8 header file name

* fix the typo

[ROCm/rccl commit: 6777e65c1d]

projects/rccl/test/common/CollectiveArgs.cpp

@@ -199,8 +199,10 @@ namespace RcclUnitTesting
         case ncclUint32:   ss << scalarsPerRank.U4[this->globalRank]; break;
         case ncclInt64:    ss << scalarsPerRank.I8[this->globalRank]; break;
         case ncclUint64:   ss << scalarsPerRank.U8[this->globalRank]; break;
+        case ncclFp8E4M3:  ss << scalarsPerRank.F1[this->globalRank]; break;
         case ncclFloat32:  ss << scalarsPerRank.F4[this->globalRank]; break;
         case ncclFloat64:  ss << scalarsPerRank.F8[this->globalRank]; break;
+        case ncclFp8E5M2:  ss << scalarsPerRank.B1[this->globalRank]; break;
         case ncclBfloat16: ss << scalarsPerRank.B2[this->globalRank]; break;
         default:           ss << "(UNKNOWN)";
         }