[CCCL] 瘦身 + 补全: 移除 cudax/python/libcudacxx-tests 冗余文件, 新增 c2h 测试助手 + cmake 构建系统 + 8 个 CUDA thrust examples

变更摘要:
- 删除: cudax/ (783 files, 7.2M) — 实验性组件，竞赛不需要
- 删除: python/ (226 files, 2.0M) — Python 绑定，竞赛不需要
- 删除: libcudacxx/{test,benchmarks,codegen,cmake,share} (4432 files, 31M)
  保留: libcudacxx/include/ (1463 headers, cuda::std 编译依赖)
- 新增: c2h/ (27 files) — CUB Catch2 测试辅助头文件，编译 243 个测试必需
- 新增: cmake/ (29 files) — CCCL 原生 CMake 构建系统
- 新增: thrust/examples/cuda/ (7 files) + cpp_integration/ (1 file)
  async_reduce, custom_temporary_allocation, explicit_cuda_stream,
  global_device_vector, range_view, unwrap_pointer, wrap_pointer, device

结果: cccl_upstream 从 74M→35M (瘦身 53%), 核心内容 100% 保留:
  27/27 tuning headers, 78 benchmarks, 243 tests,
  60 thrust examples, 18 CUB examples, 全部编译头文件

cccl_upstream/cudax/benchmarks/CMakeLists.txt

@@ -1,69 +0,0 @@
-include(${CMAKE_SOURCE_DIR}/benchmarks/cmake/CCCLBenchmarkRegistry.cmake)
-
-cccl_get_nvbench()
-cccl_get_nvbench_helper()
-
-set(benches_root "${CMAKE_CURRENT_LIST_DIR}")
-
-function(get_recursive_subdirs subdirs)
-  set(dirs)
-  file(
-    GLOB_RECURSE contents
-    CONFIGURE_DEPENDS
-    LIST_DIRECTORIES ON
-    "${CMAKE_CURRENT_LIST_DIR}/bench/*"
-  )
-
-  foreach (bench_dir IN LISTS contents)
-    if (IS_DIRECTORY "${bench_dir}")
-      list(APPEND dirs "${bench_dir}")
-    endif()
-  endforeach()
-
-  set(${subdirs} "${dirs}" PARENT_SCOPE)
-endfunction()
-
-function(add_bench target_name bench_name bench_src)
-  set(bench_target ${bench_name})
-  set(${target_name} ${bench_target} PARENT_SCOPE)
-
-  cccl_add_executable(${bench_target} SOURCES "${bench_src}")
-  target_link_libraries(
-    ${bench_target}
-    PRIVATE #
-      cccl.nvbench_helper
-      nvbench::main
-  )
-endfunction()
-
-function(add_bench_dir bench_dir)
-  file(GLOB bench_srcs CONFIGURE_DEPENDS "${bench_dir}/*.cu")
-  file(RELATIVE_PATH bench_prefix "${benches_root}" "${bench_dir}")
-  file(TO_CMAKE_PATH "${bench_prefix}" bench_prefix)
-  string(REPLACE "/" "." bench_prefix "${bench_prefix}")
-
-  foreach (bench_src IN LISTS bench_srcs)
-    get_filename_component(bench_name "${bench_src}" NAME_WLE)
-    string(PREPEND bench_name "cudax.${bench_prefix}.")
-    register_cccl_benchmark("${bench_name}" "")
-
-    string(APPEND bench_name ".base")
-
-    add_bench(base_bench_target ${bench_name} "${bench_src}")
-    target_link_libraries(${bench_name} PRIVATE cudax.compiler_interface)
-    target_compile_options(
-      ${bench_name}
-      PRIVATE
-        "$<$<COMPILE_LANG_AND_ID:CUDA,NVIDIA>:--extended-lambda>"
-        # cudax.compiler_interface enables assertions for tests/examples; benchmarks should measure release behavior.
-        "$<$<COMPILE_LANGUAGE:CUDA>:-UCCCL_ENABLE_ASSERTIONS>"
-        "$<$<COMPILE_LANGUAGE:CXX>:-UCCCL_ENABLE_ASSERTIONS>"
-    )
-  endforeach()
-endfunction()
-
-get_recursive_subdirs(subdirs)
-
-foreach (subdir IN LISTS subdirs)
-  add_bench_dir("${subdir}")
-endforeach()