[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/libcudacxx/test/support/cmpxchg_loop.h

@@ -0,0 +1,69 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, 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
+//
+//===----------------------------------------------------------------------===//
+
+#include <cuda/std/atomic>
+
+#include "test_macros.h"
+
+template <class A, class T>
+TEST_FUNC bool cmpxchg_weak_loop(A& atomic, T& expected, T desired)
+{
+  for (int i = 0; i < 10; i++)
+  {
+    if (atomic.compare_exchange_weak(expected, desired) == true)
+    {
+      return true;
+    }
+  }
+
+  return false;
+}
+
+template <class A, class T>
+TEST_FUNC bool
+cmpxchg_weak_loop(A& atomic, T& expected, T desired, cuda::std::memory_order success, cuda::std::memory_order failure)
+{
+  for (int i = 0; i < 10; i++)
+  {
+    if (atomic.compare_exchange_weak(expected, desired, success, failure) == true)
+    {
+      return true;
+    }
+  }
+
+  return false;
+}
+
+template <class A, class T>
+TEST_FUNC bool c_cmpxchg_weak_loop(A* atomic, T* expected, T desired)
+{
+  for (int i = 0; i < 10; i++)
+  {
+    if (cuda::std::atomic_compare_exchange_weak(atomic, expected, desired) == true)
+    {
+      return true;
+    }
+  }
+
+  return false;
+}
+
+template <class A, class T>
+TEST_FUNC bool
+c_cmpxchg_weak_loop(A* atomic, T* expected, T desired, cuda::std::memory_order success, cuda::std::memory_order failure)
+{
+  for (int i = 0; i < 10; i++)
+  {
+    if (cuda::std::atomic_compare_exchange_weak_explicit(atomic, expected, desired, success, failure) == true)
+    {
+      return true;
+    }
+  }
+
+  return false;
+}