[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/test/stf/reclaiming/graph.cu

@@ -1,74 +0,0 @@
-//===----------------------------------------------------------------------===//
-//
-// Part of CUDASTF in CUDA C++ Core Libraries,
-// under the Apache License v2.0 with LLVM Exceptions.
-// See https://llvm.org/LICENSE.txt for license information.
-// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-// SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES.
-//
-//===----------------------------------------------------------------------===//
-
-#include <cuda/experimental/__stf/graph/graph_ctx.cuh>
-
-#include <iostream>
-
-#if !_CCCL_COMPILER(MSVC)
-using namespace cuda::experimental::stf;
-
-__global__ void kernel()
-{
-  // No-op
-}
-#endif // !_CCCL_COMPILER(MSVC)
-
-int main([[maybe_unused]] int argc, [[maybe_unused]] char** argv)
-{
-// TODO fix setenv
-#if !_CCCL_COMPILER(MSVC)
-  int nblocks       = 4;
-  size_t block_size = 1024 * 1024;
-
-  if (argc > 1)
-  {
-    nblocks = atoi(argv[1]);
-  }
-
-  if (argc > 2)
-  {
-    block_size = atoi(argv[2]);
-  }
-
-  // At most 1 buffer is allocated at the same time
-  setenv("MAX_ALLOC_CNT", "1", 0);
-
-  graph_ctx ctx;
-
-  ::std::vector<logical_data<slice<char>>> handles(nblocks);
-
-  char* h_buffer = new char[nblocks * block_size];
-
-  for (int i = 0; i < nblocks; i++)
-  {
-    handles[i] = ctx.logical_data(make_slice(&h_buffer[i * block_size], block_size));
-    handles[i].set_symbol("D_" + std::to_string(i));
-  }
-
-  // We only 2 buffers, we are forced to reuse the buffer from D0 for D2
-  for (int i = 0; i < 3; i++)
-  {
-    ctx.task(handles[i % nblocks].rw())->*[&](cudaStream_t s, auto /*unused*/) {
-      kernel<<<1, 1, 0, s>>>();
-    };
-  }
-
-  ctx.submit();
-
-  if (argc > 3)
-  {
-    std::cout << "Generating DOT output in " << argv[3] << '\n';
-    ctx.print_to_dot(argv[1]);
-  }
-
-  ctx.finalize();
-#endif // !_CCCL_COMPILER(MSVC)
-}