Pull in r345199 from upstream libc++ trunk (by Petr Hosek):
Revert "Teach __libcpp_is_floating_point that __fp16 and _Float16 are floating-point types." This reverts commits r333103 and r333108. _Float16 and __fp16 are C11 extensions and compilers other than Clang don't define these for C++. Differential Revision: https://reviews.llvm.org/D53670
This prevents "_Float16 is not supported on this target" errors in
libc++'s type_traits header, in some cases.
Reported by: Charlie Li
MFC after: 3 days