[CCCL] Add missing CCCL components: c2h, nvbench_helper, cmake, cudax, AGENTS.md

Added 863 files from NVIDIA/cccl sparse checkout:
- c2h/ (27 files): Catch2 test helpers — generators, validators, runner
- nvbench_helper/ (10 files): Benchmark harness utilities
- cmake/ (29 files): CMake presets and build helpers
- cudax/ (794 files): Experimental CUDA extensions
- AGENTS.md: NVIDIA's official AI agent instructions for CCCL
- CMakePresets.json: Standardized build configurations
- cccl-version.json: Version tracking

Also added CCCL_ASSET_MAP.md mapping all 4295 CCCL files to
competition value and PRD items.

cccl_upstream now covers 100% of competition-critical assets:
- 27 tuning headers (SM80/90/100 benchmark data)
- 32 dispatch headers (algorithm implementations)
- 60 Thrust examples (correctness verification)
- 217 CUB Catch2 tests (regression matrix)
- 153 CUB benchmarks (parameter space search)
- 18 CUB examples (API verification)
- 27 test helpers + benchmark harness
- 794 cudax experimental extensions

cccl_upstream/cudax/examples/stf/frozen_data_init.cu

@@ -0,0 +1,53 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of CUDASTF in CUDA C++ Core Libraries,
+// under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+// SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES.
+//
+//===----------------------------------------------------------------------===//
+
+/**
+ * @file
+ *
+ * @brief Illustrate how we can use frozen data to initialize constant data
+ *
+ */
+
+#include <cuda/experimental/stf.cuh>
+
+using namespace cuda::experimental::stf;
+
+int main()
+{
+  context ctx;
+
+  /* Create a piece of data that can be use many times without further synchronizations */
+  auto buffer = ctx.logical_data(shape_of<slice<double, 2>>(128, 64)).set_symbol("buffer");
+  ctx.parallel_for(buffer.shape(), buffer.write())->*[] __device__(size_t i, size_t j, auto b) {
+    b(i, j) = sin(-1.0 * i) + cos(2.0 * j);
+  };
+
+  auto frozen_buffer = ctx.freeze(buffer);
+
+  auto h_buf = frozen_buffer.get(data_place::host()).first;
+  auto d_buf = frozen_buffer.get(data_place::current_device()).first;
+
+  cuda_safe_call(cudaStreamSynchronize(ctx.fence()));
+
+  auto lX = ctx.logical_data(buffer.shape()).set_symbol("X");
+  ctx.parallel_for(lX.shape(), lX.write()).set_symbol("X=buf")->*[d_buf] __device__(size_t i, size_t j, auto x) {
+    x(i, j) = d_buf(i, j);
+  };
+
+  ctx.parallel_for(exec_place::host(), lX.shape(), lX.read()).set_symbol("check buf")
+      ->*[h_buf](size_t i, size_t j, auto x) {
+            EXPECT(fabs(x(i, j) - h_buf(i, j)) < 0.0001);
+          };
+
+  // Make sure all tasks are done before unfreezing
+  frozen_buffer.unfreeze(ctx.fence());
+
+  ctx.finalize();
+}