[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/test/stf/reductions/redux_test2.cu

@@ -0,0 +1,78 @@
+//===----------------------------------------------------------------------===//
+//
+// 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.
+//
+//===----------------------------------------------------------------------===//
+
+#include <cuda/experimental/__stf/stream/interfaces/slice_reduction_ops.cuh>
+#include <cuda/experimental/__stf/stream/stream_ctx.cuh>
+
+using namespace cuda::experimental::stf;
+
+__global__ void add(int* ptr, int value)
+{
+  *ptr = *ptr + value;
+}
+
+__global__ void check_val(const int* ptr, int expected)
+{
+  assert(*ptr == expected);
+}
+
+/*
+ * This test ensures that we can reconstruct a piece of data where there are
+ * multiple "shared" instances, and "redux" instances
+ */
+int main()
+{
+  stream_ctx ctx;
+
+  int ndevs;
+  cuda_safe_call(cudaGetDeviceCount(&ndevs));
+
+  if (ndevs < 2)
+  {
+    fprintf(stderr, "Skipping test: need at least 2 devices.\n");
+    return 0;
+  }
+
+  auto redux_op = std::make_shared<slice_reduction_op_sum<int>>();
+
+  int a = 17;
+
+  // init op (17)
+  auto handle = ctx.logical_data(make_slice(&a, 1));
+
+  // RW dev0 (18)
+  ctx.task(exec_place::device(0), handle.rw())->*[](auto stream, auto s) {
+    add<<<1, 1, 0, stream>>>(s.data_handle(), 1);
+  };
+
+  // READ dev1 (18)
+  ctx.task(exec_place::device(1), handle.read())->*[](auto stream, auto s) {
+    check_val<<<1, 1, 0, stream>>>(s.data_handle(), 18);
+  };
+
+  // REDUX dev1 (18 + 42)
+  ctx.task(exec_place::device(1), handle.relaxed(redux_op))->*[](auto stream, auto s) {
+    add<<<1, 1, 0, stream>>>(s.data_handle(), 42);
+  };
+
+  // READ dev0 (18 + 42)
+  ctx.task(exec_place::device(0), handle.read())->*[](auto stream, auto s) {
+    check_val<<<1, 1, 0, stream>>>(s.data_handle(), 18 + 42);
+  };
+
+  // READ
+  ctx.task(exec_place::host(), handle.read())->*[](auto stream, auto s) {
+    cuda_safe_call(cudaStreamSynchronize(stream));
+    EXPECT(s(0) == 18 + 42);
+    // printf("VALUE %d expected %d\n", s(0), 18 + 42);
+  };
+
+  ctx.finalize();
+}