Bug 22345 – [internals] Make it easy to pre-define target-specific native types.

For example: - __fp16 (half-float on ARM) - __bf16 (brain-float on ARM/ARM64) - __float80 (80-bit real on x86) - __float128 (128-bit real on x86) - __ibm128 (128-bit IBM float on PPC) - __ieee128 (128-bit IEEE float on PPC) Other C11 extension types could be considered too: - _FloatN and _FloatNx - _Decimal - _Fract - _Accum - _Sat As well as an internal va_list type. These type extensions would be made available using the isSpecial enum logic that is already present in the dmd front-end, which lets the front-end know just to ignore it.

THIS ISSUE HAS BEEN MOVED TO GITHUB https://github.com/dlang/dmd/issues/19989 DO NOT COMMENT HERE ANYMORE, NOBODY WILL SEE IT, THIS ISSUE HAS BEEN MOVED TO GITHUB