[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/set_difference_descending.cu

@@ -0,0 +1,56 @@
+#include <thrust/functional.h>
+#include <thrust/set_operations.h>
+#include <thrust/sort.h>
+
+#include <unittest/unittest.h>
+
+template <typename Vector>
+void TestSetDifferenceDescendingSimple()
+{
+  using T        = typename Vector::value_type;
+  using Iterator = typename Vector::iterator;
+
+  Vector a{5, 4, 2, 0}, b{6, 4, 3, 3, 0};
+
+  Vector ref{5, 2};
+  Vector result(2);
+
+  Iterator end =
+    thrust::set_difference(a.begin(), a.end(), b.begin(), b.end(), result.begin(), ::cuda::std::greater<T>());
+
+  ASSERT_EQUAL_QUIET(result.end(), end);
+  ASSERT_EQUAL(ref, result);
+}
+DECLARE_VECTOR_UNITTEST(TestSetDifferenceDescendingSimple);
+
+template <typename T>
+void TestSetDifferenceDescending(const size_t n)
+{
+  thrust::host_vector<T> temp = unittest::random_integers<T>(2 * n);
+  thrust::host_vector<T> h_a(temp.begin(), temp.begin() + n);
+  thrust::host_vector<T> h_b(temp.begin() + n, temp.end());
+
+  thrust::sort(h_a.begin(), h_a.end(), ::cuda::std::greater<T>());
+  thrust::sort(h_b.begin(), h_b.end(), ::cuda::std::greater<T>());
+
+  thrust::device_vector<T> d_a = h_a;
+  thrust::device_vector<T> d_b = h_b;
+
+  thrust::host_vector<T> h_result(n);
+  thrust::device_vector<T> d_result(n);
+
+  typename thrust::host_vector<T>::iterator h_end;
+  typename thrust::device_vector<T>::iterator d_end;
+
+  h_end =
+    thrust::set_difference(h_a.begin(), h_a.end(), h_b.begin(), h_b.end(), h_result.begin(), ::cuda::std::greater<T>());
+  h_result.resize(h_end - h_result.begin());
+
+  d_end =
+    thrust::set_difference(d_a.begin(), d_a.end(), d_b.begin(), d_b.end(), d_result.begin(), ::cuda::std::greater<T>());
+
+  d_result.resize(d_end - d_result.begin());
+
+  ASSERT_EQUAL(h_result, d_result);
+}
+DECLARE_VARIABLE_UNITTEST(TestSetDifferenceDescending);