[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/cpp/read_const.cu

@@ -0,0 +1,93 @@
+//===----------------------------------------------------------------------===//
+//
+// 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 This test ensures that read() can be called on const logical_data
+ *        (typed and untyped).
+ */
+
+#include <cuda/experimental/stf.cuh>
+
+using namespace cuda::experimental::stf;
+
+struct foo
+{
+  // Intentionally choose an odd (actually prime) size
+  foo(context& ctx)
+  {
+    l = ctx.logical_data(shape_of<slice<int>>(50867));
+  }
+
+  void set(context& ctx, int val)
+  {
+    ctx.parallel_for(l.shape(), l.write())->*[=] _CCCL_DEVICE(size_t i, auto dl) {
+      dl(i) = val;
+    };
+  }
+
+  void copy_from(context& ctx, const foo& other)
+  {
+    ctx.parallel_for(l.shape(), l.write(), other.l.read())->*[=] _CCCL_DEVICE(size_t i, auto dl, auto dotherl) {
+      dl(i) = dotherl(i);
+    };
+  }
+
+  void ensure(context& ctx, int val)
+  {
+    std::ignore = val;
+    ctx.parallel_for(l.shape(), l.read())->*[=] _CCCL_DEVICE(size_t i, auto dl) {
+      assert(dl(i) == val);
+    };
+  }
+
+  auto& get_l() const
+  {
+    return l;
+  }
+
+  logical_data<slice<int>> l;
+};
+
+void read_only_access(context& ctx, const foo& f)
+{
+  ctx.parallel_for(f.l.shape(), f.l.read())->*[] _CCCL_DEVICE(size_t i, auto dl) {
+    // no-op
+  };
+
+  ctx.parallel_for(f.get_l().shape(), f.get_l().read())->*[] _CCCL_DEVICE(size_t i, auto dl) {
+    // no-op
+  };
+}
+
+void read_only_access_untyped(const logical_data_untyped& ld)
+{
+  auto dep = ld.read();
+  (void) dep;
+}
+
+int main()
+{
+  context ctx;
+
+  foo A(ctx);
+  A.set(ctx, 42);
+  A.ensure(ctx, 42);
+  foo B(ctx);
+  B.copy_from(ctx, A);
+  B.ensure(ctx, 42);
+
+  read_only_access(ctx, A);
+
+  const logical_data_untyped& ld_untyped = A.l;
+  read_only_access_untyped(ld_untyped);
+
+  ctx.finalize();
+}