updates based on PR feedback

There are two major changes (and a few minor ones which are not listed here...see PR discussion for details)

1. Eigen::half implementations for HIP and CUDA have been merged.
This means that
- `CUDA/Half.h` and `HIP/hcc/Half.h` got merged to a new file `GPU/Half.h`
- `CUDA/PacketMathHalf.h` and `HIP/hcc/PacketMathHalf.h` got merged to a new file `GPU/PacketMathHalf.h`
- `CUDA/TypeCasting.h` and `HIP/hcc/TypeCasting.h` got merged to a new file `GPU/TypeCasting.h`

After this change the `HIP/hcc` directory only contains one file `math_constants.h`. That will go away too once that file becomes a part of the HIP install.

2. new macros EIGEN_GPUCC, EIGEN_GPU_COMPILE_PHASE and EIGEN_HAS_GPU_FP16 have been added and the code has been updated to use them where appropriate.
- `EIGEN_GPUCC` is the same as `(EIGEN_CUDACC || EIGEN_HIPCC)`
- `EIGEN_GPU_DEVICE_COMPILE` is the same as `(EIGEN_CUDA_ARCH || EIGEN_HIP_DEVICE_COMPILE)`
- `EIGEN_HAS_GPU_FP16` is the same as `(EIGEN_HAS_CUDA_FP16 or EIGEN_HAS_HIP_FP16)`

Eigen/src/Core/GenericPacketMath.h

@@ -301,13 +301,11 @@ template<typename Scalar, typename Packet> EIGEN_DEVICE_FUNC inline void pstoreu
  { pstore(to, from); }
 
 /** \internal tries to do cache prefetching of \a addr */
-template<typename Scalar>
-  #if !defined(EIGEN_HIPCC)
-  EIGEN_DEVICE_FUNC
-  #endif
-  inline void prefetch(const Scalar* addr)
+template<typename Scalar> EIGEN_DEVICE_FUNC inline void prefetch(const Scalar* addr)
 {
-#ifdef EIGEN_CUDA_ARCH
+#if defined(EIGEN_HIP_DEVICE_COMPILE)
+  // do nothing
+#elif defined(EIGEN_CUDA_ARCH)
 #if defined(__LP64__)
   // 64-bit pointer operand constraint for inlined asm
   asm(" prefetch.L1 [ %1 ];" : "=l"(addr) : "l"(addr));
@@ -534,7 +532,7 @@ inline void palign(PacketType& first, const PacketType& second)
 ***************************************************************************/
 
 // Eigen+CUDA does not support complexes.
-#if !defined(EIGEN_CUDACC) && !defined(EIGEN_HIPCC)
+#if !defined(EIGEN_GPUCC)
 
 template<> inline std::complex<float> pmul(const std::complex<float>& a, const std::complex<float>& b)
 { return std::complex<float>(real(a)*real(b) - imag(a)*imag(b), imag(a)*real(b) + real(a)*imag(b)); }