[clang] [WIP][CLANG][AArch64] Add the modal 8 bit floating-point scalar type (PR #97277)

[via cfe-commits]

CarolineConcatto wrote:

Just in I don't believe this patch is ready, but I would like to share so I could get some pointers of what is missing.
I  also have  question about implementation:
Should we convert to a)unsigned 8 bits correspondent in CodeGenTypes.cpp or b)should we create another opaque  in LLVM  or c) something else?
In the implementation should we use  UNSIGNED_TYPE(Fpm8, Fpm8Ty) in BuiltinTypes.def or needs to adds OPAQUE?

https://github.com/llvm/llvm-project/pull/97277