[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/thrust/benchmarks/bench/all_of/basic.cu

@@ -0,0 +1,41 @@
+// SPDX-FileCopyrightText: Copyright (c) 2026, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+#include <thrust/device_vector.h>
+#include <thrust/fill.h>
+#include <thrust/logical.h>
+
+#include <cuda/functional>
+#include <cuda/memory_pool>
+#include <cuda/stream>
+
+#include "nvbench_helper.cuh"
+
+template <typename T>
+static void basic(nvbench::state& state, nvbench::type_list<T>)
+{
+  T val = 1;
+  // set up input
+  const auto elements       = static_cast<std::size_t>(state.get_int64("Elements"));
+  const auto common_prefix  = state.get_float64("MismatchAt");
+  const auto mismatch_point = static_cast<std::size_t>(static_cast<double>(elements) * common_prefix);
+
+  thrust::device_vector<T> dinput(elements, thrust::no_init);
+  thrust::fill(dinput.begin(), dinput.begin() + mismatch_point, T{0});
+  thrust::fill(dinput.begin() + mismatch_point, dinput.end(), val);
+
+  state.add_global_memory_reads<T>(mismatch_point + 1);
+  state.add_global_memory_writes<size_t>(1);
+
+  caching_allocator_t alloc{};
+
+  state.exec(
+    nvbench::exec_tag::gpu | nvbench::exec_tag::no_batch | nvbench::exec_tag::sync, [&](nvbench::launch& launch) {
+      do_not_optimize(thrust::all_of(policy(alloc, launch), dinput.begin(), dinput.end(), cuda::equal_to_value{val}));
+    });
+}
+
+NVBENCH_BENCH_TYPES(basic, NVBENCH_TYPE_AXES(fundamental_types))
+  .set_name("base")
+  .add_int64_power_of_two_axis("Elements", nvbench::range(16, 28, 4))
+  .add_float64_axis("MismatchAt", std::vector{1.0, 0.5, 0.01});