TL;DR: Add an experimental FSR Ray Regeneration backend on Windows, run the K

network natively on HIP with Vulkan->HIP shared memory, extend K kernels to
gfx11 (RDNA3), and add dependency/packaging/test tooling around them.

Windows: FSR RR backend, native HIP K runner, gfx11 K kernels, tooling

FSR Ray Regeneration (experimental)
- Add fsr_rr_backend.{h,cpp}, fsr_rr_passes.h and the fsr_rr/ prepare and
  compose compute shaders. They run AMD's FidelityFX denoiser
  (amd_fidelityfx_denoiser_dx12.dll, SDK 2.3, not bundled) on a private
  vkd3d-proton device.
- Add the amdxc64 shim (tools/windows/amdxc64_shim) and two vkd3d-proton
  patches for libcxx includes. AMD's amdxc64.dll is never loaded.
- d4r_nvngx_windows: RrFeature now selects between the native DLSS RR
  backend and the FSR backend (D4R_RR_BACKEND=fsr). After the denoiser it
  runs an internal Super Sampling stage (feature 1) to upscale. If SR is
  unavailable and output != render size, the frame fails; it is never
  returned noisy.
- d4r.ini / player_config: new [RayReconstruction] FsrDenoiser,
  FsrStabilityBias (auto or >= 0) and FsrDetail (0-1), with validation, and
  matching D4R_FSR_* environment variables.
- Add the FSR RR fixture, output-check and ngx test scripts, and a shader
  compile script.

Native HIP K runner (Windows)
- CMake: add d4r_engine_k_hip (d4r-k-hip) plus separate volk/engine copies,
  because VK_USE_PLATFORM_WIN32_KHR changes volk's entry points.
- Share Vulkan memory with HIP via Win32 opaque handles on Windows
  (VK_KHR_external_memory_win32) and via fds on Linux. Handle types are now
  intptr_t, and the platform choice is in D4R_EXT_MEMORY_* macros.
- HipNetwork: add debugRepeatHashes and debugNodeMicros, exposed as
  D4R_HIP_REPEAT_CHECK, D4R_HIP_REPEAT_ALL and D4R_HIP_NODE_PROFILE in
  replay.cpp.
- replay.cpp: display-resolution-motion frames (FrameFlagDisplayMotion) no
  longer require near2rnd == out2rnd. Both values must still be in (0, 1].

K kernels on gfx11
- Allow position-token and enc1 tiling on __GFX11__ as well as gfx12.
- Add an opt-in ds_bpermute cross-half exchange
  (D4R_WINDOWS_GFX11_DS_EXCHANGE, enabled for K on Windows gfx11). M keeps
  its LDS path.
- Fix a block-barrier race in the position layer. Without POS_VSHARE a fast
  wave could overwrite its hb rows while a slower wave still read them for a
  later head, giving rare wrong results on gfx1101 (enc0 outputs differing
  between identical launches).

Packaging and install
- install.bat, d4r.ini, WINDOWS_README: FSR option, and
  install_dependencies.py to install pinned NVIDIA dependencies with staged,
  hash-verified replacement and rollback.
- Update the release package, local package refresh and test scripts.
- Add the InteropBatchInputCopies option (default false). Batching is faster
  on some synthetic runs, but player temporal exactness is unresolved.

Tooling and tests
- K model: capture_k_weights, build_k_model, compare_k_outputs,
  benchmark_k_hip, build_capture_runtime, test_k_model and the
  zluda-k-output-capture patch. Captured weights and models are private
  local assets and must not enter release archives.
- Synthetic probes: test_m_ngx, test_sr_dynamic,
  prepare_rr_dynamic_fixture and the d3d12/rr_ngx probe updates.
- Add tests/windows/test_install_dependencies.py.

Docs
- Update the windows-fsr-rr, k-performance, local-package, rr-interop and
  rr-model-cache docs. Add windows-parity-handoff.md and
  windows-parity-continuation-prompt.md.

tools/windows/ngx_windows_runtime.h

@@ -14,6 +14,7 @@
 #include <atomic>
 #include <memory>
 #include <mutex>
+#include <array>
 
 namespace d4r::win {
 inline std::string env_path(const char* name) {
@@ -312,6 +313,34 @@ public:
         std::lock_guard<std::mutex> lock(rt_->mutex); cleanup();
     }
     Feature(const Feature&) = delete;
+    // Query the loaded SR library, using NVIDIA's optimal-settings callback contract.
+    // An unavailable/invalid query keeps the conservative creation-size-only range.
+    std::array<unsigned, 4> render_limits() {
+        std::lock_guard<std::mutex> lock(rt_->mutex);
+        rt_->current();
+        std::array<unsigned, 4> limits{width, height, width, height};
+        void* caps = nullptr;
+        if (rt_->capabilities(&caps) != 1 || !caps) return limits;
+        struct Guard { Runtime& rt; void* p; ~Guard() { (void)rt.destroy(p); } } guard{*rt_, caps};
+        void* callback = nullptr;
+        if (d4r_ngx_get_void(caps, "DLSSOptimalSettingsCallback", &callback) != 1 || !callback) return limits;
+        d4r_ngx_set_uint(caps, "Width", ngx::uint_value(parameters_, "OutWidth"));
+        d4r_ngx_set_uint(caps, "Height", ngx::uint_value(parameters_, "OutHeight"));
+        d4r_ngx_set_int(caps, "PerfQualityValue", ngx::int_value(parameters_, "PerfQualityValue"));
+        d4r_ngx_set_int(caps, "RTXValue", 0);
+        using Optimal = unsigned(*)(void*);
+        if (reinterpret_cast<Optimal>(callback)(caps) != 1) return limits;
+        const unsigned optimalWidth = ngx::uint_value(caps, "OutWidth");
+        const unsigned optimalHeight = ngx::uint_value(caps, "OutHeight");
+        const std::array<unsigned, 4> queried{
+            ngx::uint_value(caps, "DLSS.Get.Dynamic.Min.Render.Width", optimalWidth),
+            ngx::uint_value(caps, "DLSS.Get.Dynamic.Min.Render.Height", optimalHeight),
+            ngx::uint_value(caps, "DLSS.Get.Dynamic.Max.Render.Width", optimalWidth),
+            ngx::uint_value(caps, "DLSS.Get.Dynamic.Max.Render.Height", optimalHeight)};
+        if (queried[0] && queried[1] && queried[2] >= queried[0] && queried[3] >= queried[1]) limits = queried;
+        std::fprintf(stderr, "D4R_SR_RENDER_LIMITS min=%ux%u max=%ux%u\n", limits[0], limits[1], limits[2], limits[3]);
+        return limits;
+    }
     void drain() {
         if (!std::getenv("D4R_ASYNC_INTEROP")) return;
         wait_worker();