[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/cmake/ThrustRunExample.cmake

@@ -0,0 +1,52 @@
+# Inputs:
+#
+# Variable             | Type     | Doc
+# ---------------------|----------|--------------------------------------
+# EXAMPLE_EXECUTABLE   | FilePath | Path to example executable
+# FILECHECK_ENABLED    | Boolean  | Run FileCheck comparison test
+# FILECHECK_EXECUTABLE | FilePath | Path to the LLVM FileCheck utility
+# REFERENCE_FILE       | FilePath | Path to the FileCheck reference file
+
+if (FILECHECK_ENABLED)
+  if (NOT EXISTS "${REFERENCE_FILE}")
+    message(
+      FATAL_ERROR
+      "FileCheck requested for '${EXAMPLE_EXECUTABLE}', but reference file "
+      "does not exist at '${REFERENCE_FILE}`."
+    )
+  endif()
+
+  # If the reference file is empty, validate that the example doesn't
+  # produce any output.
+  file(SIZE "${REFERENCE_FILE}" file_size)
+  message("${REFERENCE_FILE}: ${file_size} bytes")
+
+  if (file_size EQUAL 0)
+    set(check_empty_output TRUE)
+    set(filecheck_command)
+  else()
+    set(check_empty_output FALSE)
+    set(filecheck_command COMMAND "${FILECHECK_EXECUTABLE}" "${REFERENCE_FILE}")
+  endif()
+endif()
+
+execute_process(
+  COMMAND "${EXAMPLE_EXECUTABLE}" ${filecheck_command}
+  RESULT_VARIABLE exit_code
+  OUTPUT_VARIABLE stdout
+  ERROR_VARIABLE stderr
+)
+
+if (NOT 0 EQUAL exit_code)
+  message(FATAL_ERROR "${EXAMPLE_EXECUTABLE} failed (${exit_code}):\n${stderr}")
+endif()
+
+if (check_empty_output)
+  string(LENGTH "${stdout}" stdout_size)
+  if (NOT stdout_size EQUAL 0)
+    message(
+      FATAL_ERROR
+      "${EXAMPLE_EXECUTABLE}: output received, but not expected:\n${stdout}"
+    )
+  endif()
+endif()