feat: add d4r inference engines and Windows parity tooling

Add native K/M/RR inference engines, HIP kernels, replay and validation
harnesses, and Windows DLSS setup and runtime support. Include GPU
profiling tools, build and packaging scripts, platform integration
patches, and implementation documentation.

kernels/k/dltss_pwin_dec1_layer.hip

@@ -0,0 +1,6 @@
+// CU mode: the waves of a window share one L0 cache (1-6% faster here; slower in the wide layers)
+// QKV chains two at a time (K-race port): -1.7% on top of op_imgt (paired gfx1200 replay, exact)
+// d4r-build-flags: -mcumode -DPWIN_QKV_SPLIT
+#define PWIN_F32ACC
+#include "pwin_layer.h"
+PWIN_DECODER(dltss_pwin_dec1_layer, 2, 64, 64)