[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/stream_per_thread.cmake

@@ -0,0 +1,13 @@
+# This test should always use per-thread streams on NVCC.
+set_target_properties(
+  ${test_target}
+  PROPERTIES
+    COMPILE_OPTIONS
+      $<$<COMPILE_LANG_AND_ID:CUDA,NVIDIA>:--default-stream=per-thread>
+)
+
+# NVC++ does not have an equivalent option, and will always
+# use the global stream by default.
+if (CMAKE_CUDA_COMPILER_ID STREQUAL "NVHPC")
+  set_tests_properties(${test_target} PROPERTIES WILL_FAIL ON)
+endif()