[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/unittest_tester.cu

@@ -0,0 +1,46 @@
+#include <unittest/unittest.h>
+
+void TestAssertEqual()
+{
+  ASSERT_EQUAL(0, 0);
+  ASSERT_EQUAL(1, 1);
+  ASSERT_EQUAL(-15.0f, -15.0f);
+}
+DECLARE_UNITTEST(TestAssertEqual);
+
+void TestAssertLEqual()
+{
+  ASSERT_LEQUAL(0, 1);
+  ASSERT_LEQUAL(0, 0);
+}
+DECLARE_UNITTEST(TestAssertLEqual);
+
+void TestAssertGEqual()
+{
+  ASSERT_GEQUAL(1, 0);
+  ASSERT_GEQUAL(0, 0);
+}
+DECLARE_UNITTEST(TestAssertGEqual);
+
+void TestAssertLess()
+{
+  ASSERT_LESS(0, 1);
+}
+DECLARE_UNITTEST(TestAssertLess);
+
+void TestAssertGreater()
+{
+  ASSERT_GREATER(1, 0);
+}
+DECLARE_UNITTEST(TestAssertGreater);
+
+void TestTypeName()
+{
+  ASSERT_EQUAL(unittest::type_name<char>(), "char");
+  ASSERT_EQUAL(unittest::type_name<signed char>(), "signed char");
+  ASSERT_EQUAL(unittest::type_name<unsigned char>(), "unsigned char");
+  ASSERT_EQUAL(unittest::type_name<int>(), "int");
+  ASSERT_EQUAL(unittest::type_name<float>(), "float");
+  ASSERT_EQUAL(unittest::type_name<double>(), "double");
+}
+DECLARE_UNITTEST(TestTypeName);