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/gpu_target_probe.cpp

@@ -0,0 +1,31 @@
+#include "gpu_target.h"
+#include <cstdio>
+
+int main(int argc, char** argv) {
+    using namespace d4r::diag;
+    try {
+        unsigned rejected = 0, objects = 0;
+        for (const char* target : {"gfx1200", "gfx1201"}) {
+            require_gpu_target(target, target);
+            require_gpu_target(std::string(target) + ":xnack-:sramecc-", target);
+            for (const char* invalid : {"gfx1101", "gfx12000", "gfx1202", "gfx1250", ""}) {
+                try { require_gpu_target(invalid, target); }
+                catch (const std::exception&) { ++rejected; continue; }
+                throw std::runtime_error("Invalid GPU target accepted");
+            }
+        }
+        for (int i = 1; i < argc; ++i) {
+            require_code_object_target(argv[i], D4R_TARGET_ARCH);
+            const auto other = std::string(D4R_TARGET_ARCH) == "gfx1201" ? "gfx1200" : "gfx1201";
+            try { require_code_object_target(argv[i], other); }
+            catch (const std::exception&) { ++rejected; ++objects; continue; }
+            throw std::runtime_error("Wrong code-object target accepted");
+        }
+        if (!objects) throw std::runtime_error("Supply actual compiled code objects");
+        std::printf("PASS GPU_TARGET_CONTRACT target=%s objects=%u rejected=%u hardware_executed=0\n",
+            D4R_TARGET_ARCH, objects, rejected);
+        return 0;
+    } catch (const std::exception& e) {
+        std::fprintf(stderr, "FAIL GPU_TARGET_CONTRACT %s\n", e.what()); return 5;
+    }
+}