[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/internal/test/thrust.example.discrete_voronoi.filecheck

@@ -0,0 +1,11 @@
+     CHECK: [Initialize {{[0-9]+}}x{{[0-9]+}} Image]
+CHECK-NEXT:   ( {{[0-9.]+}}ms )
+CHECK-NEXT: [Copy to Device]
+CHECK-NEXT:   ( {{[0-9.]+}}ms )
+CHECK-NEXT: [JFA stepping]
+CHECK-NEXT:   ( {{[0-9.]+}}ms )
+CHECK-NEXT:   ( {{[0-9.]+}} MPixel/s )
+CHECK-NEXT: [Device to Host Copy]
+CHECK-NEXT:   ( {{[0-9.]+}}ms )
+CHECK-NEXT: [PGM Export]
+CHECK-NEXT:   ( {{[0-9.]+}}ms )