[clang] [AMDGPU][Clang] Support for type inferring extended image builtins for AMDGPU (PR #164358)

[Juan Manuel Martinez Caamaño via cfe-commits]

================
@@ -3,7 +3,7 @@
 
 typedef int int4 __attribute__((ext_vector_type(4)));
 typedef float float4 __attribute__((ext_vector_type(4)));
-typedef half half4 __attribute__((ext_vector_type(4)));
+typedef _Float16 half4 __attribute__((ext_vector_type(4)));
----------------
jmmartinez wrote:

`__bf16` would be a different thing since the underlying representation is not the same. 

For `__fp16/_Float16`, I'm tempted to add a new case which would be language sensitive: becomes `__fp16` for OpenCL, otherwise defaults to `_Float16`.

https://github.com/llvm/llvm-project/pull/164358