[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/asan_testing.h

@@ -0,0 +1,35 @@
+//===----------------------------------------------------------------------===//
+//
+// 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
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef ASAN_TESTING_H
+#define ASAN_TESTING_H
+
+#include "test_macros.h"
+
+#if _CCCL_HAS_FEATURE(address_sanitizer)
+extern "C" int __sanitizer_verify_contiguous_container(const void* beg, const void* mid, const void* end);
+
+template <typename T, typename Alloc>
+bool is_contiguous_container_asan_correct(const std::vector<T, Alloc>& c)
+{
+  if (std::is_same<Alloc, std::allocator<T>>::value && c.data() != nullptr)
+  {
+    return __sanitizer_verify_contiguous_container(c.data(), c.data() + c.size(), c.data() + c.capacity()) != 0;
+  }
+  return true;
+}
+
+#else
+template <typename T, typename Alloc>
+bool is_contiguous_container_asan_correct(const std::vector<T, Alloc>&)
+{
+  return true;
+}
+#endif
+
+#endif // ASAN_TESTING_H