[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/CMakeLists.txt

@@ -0,0 +1,29 @@
+foreach (thrust_target IN LISTS THRUST_TARGETS)
+  thrust_get_target_property(config_device ${thrust_target} DEVICE)
+  thrust_get_target_property(config_prefix ${thrust_target} PREFIX)
+
+  set(framework_target ${config_prefix}.test.framework)
+
+  if ("CUDA" STREQUAL "${config_device}")
+    set(
+      framework_srcs #
+      testframework.cu
+      cuda/testframework.cu
+    )
+  else()
+    # Wrap the cu file inside a .cpp file for non-CUDA builds
+    thrust_wrap_cu_in_cpp(framework_srcs testframework.cu ${thrust_target})
+  endif()
+
+  add_library(${framework_target} STATIC ${framework_srcs})
+  cccl_configure_target(${framework_target})
+  cccl_ensure_metatargets(${framework_target})
+  target_link_libraries(${framework_target} PUBLIC ${thrust_target})
+  target_include_directories(
+    ${framework_target}
+    PRIVATE "${Thrust_SOURCE_DIR}/testing"
+  )
+  if ("CUDA" STREQUAL "${config_device}")
+    thrust_configure_cuda_target(${framework_target} RDC ${THRUST_FORCE_RDC})
+  endif()
+endforeach()