[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/cudax/include/cuda/experimental/group.cuh

@@ -0,0 +1,41 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of CUDA Experimental 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) 2026 NVIDIA CORPORATION & AFFILIATES.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef _CUDA_EXPERIMENTAL_GROUP
+#define _CUDA_EXPERIMENTAL_GROUP
+
+#include <cuda/std/detail/__config>
+
+#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
+#  pragma GCC system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
+#  pragma clang system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
+#  pragma system_header
+#endif // no system header
+
+#include <cuda/experimental/__group/concepts.cuh>
+#include <cuda/experimental/__group/fwd.cuh>
+#include <cuda/experimental/__group/group.cuh>
+#include <cuda/experimental/__group/implicit_hierarchy.cuh>
+#include <cuda/experimental/__group/invoke_one.cuh>
+#include <cuda/experimental/__group/mapping/binary_partition.cuh>
+#include <cuda/experimental/__group/mapping/composite_mapping.cuh>
+#include <cuda/experimental/__group/mapping/group_as.cuh>
+#include <cuda/experimental/__group/mapping/group_by.cuh>
+#include <cuda/experimental/__group/mapping/identity_mapping.cuh>
+#include <cuda/experimental/__group/mapping/take.cuh>
+#include <cuda/experimental/__group/queries.cuh>
+#include <cuda/experimental/__group/synchronizer/barrier_synchronizer.cuh>
+#include <cuda/experimental/__group/synchronizer/lane_synchronizer.cuh>
+#include <cuda/experimental/__group/this_group.cuh>
+#include <cuda/experimental/__group/traits.cuh>
+
+#endif // _CUDA_EXPERIMENTAL_GROUP