3.6.0 update (#2005)

* 3.6.0 update

* doc and swap stuff

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
Co-authored-by: Haicheng Wu <haichengw@nvidia.com>

include/cutlass/numeric_size.h

@@ -68,6 +68,15 @@ bits_to_bytes(T bits) {
   return (R(bits) + R(7)) / R(8);
 }
 
+/// Returns the number of bits required to hold a specified number of bytes
+template <class R = int, class T>
+CUTLASS_HOST_DEVICE
+constexpr
+R
+bytes_to_bits(T bytes) {
+  return R(bytes) * R(8);
+}
+
 /////////////////////////////////////////////////////////////////////////////////////////////////
 
 template <class T>