Build native Windows packages for both RDNA4 targets

Add explicit gfx1200/gfx1201 builds, runtime selection and per-target
code objects. Guard staging, packaging and installation against mixed
architectures; retain gfx1201 compatibility and strict K/M arithmetic.

Adapt realdody's target configuration and MSVC binary16/BOOL portability
fixes from ae530b0. Do not import its forced gfx12 layout on RDNA3.
Add exhaustive binary16 and C++/PowerShell target contract regressions.

Validation: 20 CTest gates on RX 9070 XT; four software-only gfx1200 gates;
MSVC portability compile checks; exact 4K K async and M control outputs;
isolated package install/restore and mixed-target rejection tests.
No physical gfx1200 validation is claimed.

Co-authored-by: realdody <[email protected]>

tools/windows/k_layer_validate.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-"""Compare real gfx1201 K enc1/enc2 launches with the existing numpy reference."""
+"""Compare RDNA4 K enc1/enc2 launches with the existing numpy reference."""
 import argparse
 import math
 import pathlib
@@ -73,7 +73,7 @@ def main():
     # after each K stage has been compared on captured DLSS weights/activations.
     if full_abs > 0.001 or full_psnr < 60.0 or merged_abs > 0.001 or merged_psnr < 60.0:
         raise RuntimeError(f"K {args.kernel_name} numerical mismatch against pwin_model.py")
-    print(f"PASS K_REFERENCE architecture=gfx1201 kernel={args.kernel_name} "
+    print(f"PASS K_REFERENCE reference_layout=gfx12 kernel={args.kernel_name} "
           "nonzero_weights=1 transformer_executed=1 nan_inf=0")