CUTLASS 3.2.1 (#1113)

* Updates for 3.2.1 release.

* Minor fix in gemm op profiler for raster order.

* Add scheduler mapping for raster order in the kernels.

include/cutlass/half.h

@@ -33,6 +33,17 @@
     \brief Defines a class for using IEEE half-precision floating-point types in host or
       device code.
 */
+
+/*
+  Note:  CUTLASS 3x increases the host compiler requirements to C++17. However, certain
+         existing integrations of CUTLASS require C++11 host compilers.
+
+         Until this requirement can be lifted, certain headers with this annotation are required
+         to be remain consistent with C++11 syntax.
+
+         C++11 compatibility is enforced by `cutlass_test_unit_core_cpp11`.
+*/
+
 #pragma once
 
 #ifndef CUTLASS_ENABLE_F16C
@@ -158,7 +169,6 @@ public:
 
 ///////////////////////////////////////////////////////////////////////////////////////////////////
 
-
 namespace cutlass {
 
 ///////////////////////////////////////////////////////////////////////////////////////////////////