data models and 64 bit on IBM XL C

IBM XL C defines __64BIT__ if the compilation target is 64 bit: https://www.ibm.com/docs/en/xl-c-aix/13.1.2?topic=descriptions-q32-q64

And I'm not sure if it belongs into DataModels.md too but compilers targeting Windows will always define _WIN64 if compiling for any kind of 64 bit architecture.

Of interest for DataModels.md might also be the __SIZEOF_* macros defined by GCC:

__SIZEOF_INT__
__SIZEOF_LONG__
__SIZEOF_LONG_LONG__
__SIZEOF_SHORT__
__SIZEOF_POINTER__
__SIZEOF_FLOAT__
__SIZEOF_DOUBLE__
__SIZEOF_LONG_DOUBLE__
__SIZEOF_SIZE_T__
__SIZEOF_WCHAR_T__
__SIZEOF_WINT_T__
__SIZEOF_PTRDIFF_T__

Here are the size values for me on x86 32 and 64 bit Linux:

$ gcc -m64 main.c && ./a.out 
__SIZEOF_INT__         == 4
__SIZEOF_LONG__        == 8
__SIZEOF_LONG_LONG__   == 8
__SIZEOF_SHORT__       == 2
__SIZEOF_POINTER__     == 8
__SIZEOF_FLOAT__       == 4
__SIZEOF_DOUBLE__      == 8
__SIZEOF_LONG_DOUBLE__ == 16
__SIZEOF_SIZE_T__      == 8
__SIZEOF_WCHAR_T__     == 4
__SIZEOF_WINT_T__      == 4
__SIZEOF_PTRDIFF_T__   == 8

$ gcc -m32 main.c && ./a.out 
__SIZEOF_INT__         == 4
__SIZEOF_LONG__        == 4
__SIZEOF_LONG_LONG__   == 8
__SIZEOF_SHORT__       == 2
__SIZEOF_POINTER__     == 4
__SIZEOF_FLOAT__       == 4
__SIZEOF_DOUBLE__      == 8
__SIZEOF_LONG_DOUBLE__ == 12
__SIZEOF_SIZE_T__      == 4
__SIZEOF_WCHAR_T__     == 4
__SIZEOF_WINT_T__      == 4
__SIZEOF_PTRDIFF_T__   == 4

Start with DataModels.md and compare its existing coverage with the IBM XL C 64BIT, Windows _WIN64, and GCC _SIZEOF* macros mentioned in the issue. Verify the documented values and platform conditions using the linked IBM documentation and reproducible GCC 32-bit and 64-bit builds. Done means the relevant macros and data-model details are accurately documented with clear scope.