[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/testing/cuda/complex.cu

@@ -0,0 +1,24 @@
+#include <cuda_fp16.h>
+
+#include <thrust/complex.h>
+#include <thrust/detail/alignment.h>
+#include <thrust/detail/preprocessor.h>
+
+#include <unittest/unittest.h>
+
+template <typename T, typename VectorT>
+void TestComplexAlignment()
+{
+  static_assert(sizeof(thrust::complex<T>) == sizeof(VectorT));
+  static_assert(alignof(thrust::complex<T>) == alignof(VectorT));
+
+  static_assert(sizeof(thrust::complex<T const>) == sizeof(VectorT));
+  static_assert(alignof(thrust::complex<T const>) == alignof(VectorT));
+}
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<char, char2>), TestComplexCharAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<short, short2>), TestComplexShortAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<int, int2>), TestComplexIntAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<long, long2>), TestComplexLongAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<__half, __half2>), TestComplexHalfAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<float, float2>), TestComplexFloatAlignment);
+DECLARE_UNITTEST_WITH_NAME(THRUST_PP_EXPAND_ARGS(TestComplexAlignment<double, double2>), TestComplexDoubleAlignment);