[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/cub/benchmarks/bench/segmented_reduce/sum.cu

@@ -0,0 +1,16 @@
+// SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
+
+#include <nvbench_helper.cuh>
+
+// %RANGE% TUNE_ITEMS_PER_VEC_LOAD_POW2 ipv 1:2:1
+// %RANGE% TUNE_S_THREADS_PER_WARP stpw 1:32:1
+// %RANGE% TUNE_M_THREADS_PER_WARP mtpw 1:32:1
+// %RANGE% TUNE_L_NOMINAL_4B_THREADS_PER_BLOCK ltpb 128:1024:32
+// %RANGE% TUNE_S_NOMINAL_4B_ITEMS_PER_THREAD sipt 1:32:1
+// %RANGE% TUNE_M_NOMINAL_4B_ITEMS_PER_THREAD mipt 1:32:1
+// %RANGE% TUNE_L_NOMINAL_4B_ITEMS_PER_THREAD lipt 7:24:1
+
+using value_types = all_types;
+using op_t        = ::cuda::std::plus<>;
+#include "base.cuh"