[INFRA] Import NVIDIA/CCCL upstream as optimization reference library

CCCL (CUDA C++ Core Libraries) provides:
- CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk)
- Thrust: high-level parallel algorithms (transform_reduce, sort, scan)
- libcudacxx: CUDA C++ standard library (atomics, barriers, memory)
- cudax: experimental features (memory resources, allocators)
- Tuning policies: per-SM hardware-specific algorithm parameters

Competition optimization vectors mapped to CCCL:
- Output TPS (83% weight): warp_reduce, block_reduce, device_topk
- Input TPS (14% weight): device_scan, block_load, prefetch
- Cache TPS (3% weight): prefix caching strategy patterns
- Memory (0.9 util): pooled/cached/buddy allocators

Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only)
License: Apache-2.0

cccl_upstream/examples/cudax_stf/simple_stf.cu

@@ -0,0 +1,40 @@
+//===----------------------------------------------------------------------===//
+//
+// 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.cuh>
+
+#include <cstdio>
+
+using namespace cuda::experimental::stf;
+
+int main()
+{
+  context ctx;
+
+  int array[128];
+  for (size_t i = 0; i < 128; i++)
+  {
+    array[i] = i;
+  }
+
+  auto A = ctx.logical_data(array);
+
+  ctx.parallel_for(A.shape(), A.rw())->*[] __device__(size_t i, auto a) {
+    a(i) += 4;
+  };
+
+  ctx.finalize();
+
+  for (size_t i = 0; i < 128; i++)
+  {
+    printf("array[%ld] = %d\n", i, array[i]);
+  }
+  return 0;
+}