[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/examples/stdexec_stream.cu

@@ -1,82 +0,0 @@
-//===----------------------------------------------------------------------===//
-//
-// Part of CUDA Experimental 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) 2024 NVIDIA CORPORATION & AFFILIATES.
-//
-//===----------------------------------------------------------------------===//
-
-#include <cuda/experimental/execution.cuh>
-
-#include <nv/target>
-
-#include <cstdio>
-
-#include <cuda_runtime_api.h>
-
-namespace cudax = cuda::experimental;
-namespace ex    = cudax::execution;
-
-// This example demonstrates how to use the experimental CUDA implementation of
-// C++26's std::execution async tasking framework.
-
-int main()
-{
-  try
-  {
-    auto tctx = ex::thread_context{};
-    auto sctx = ex::stream_context{cuda::device_ref{0}};
-    auto gpu  = sctx.get_scheduler();
-
-    const auto bulk_shape = 10;
-    const auto bulk_fn    = [] __device__(const int index, int i) noexcept {
-      const int tid = static_cast<int>(blockIdx.x * blockDim.x + threadIdx.x);
-      if (tid < bulk_shape)
-      {
-        printf("Hello from bulk task on device! index = %d, i = %d\n", index, i);
-      }
-    };
-
-    auto start =
-      // begin work on the GPU:
-      ex::schedule(gpu)
-
-      // execute a device lambda on the GPU:
-      | ex::then([] __device__() noexcept -> int {
-          printf("Hello from lambda on device!\n");
-          return 42;
-        })
-
-      // do some parallel work on the GPU:
-      | ex::bulk(ex::par, bulk_shape, bulk_fn) //
-
-      // transfer execution back to the CPU:
-      | ex::continues_on(tctx.get_scheduler())
-
-      // execute a host/device lambda on the CPU:
-      | ex::then([] __host__ __device__(int i) noexcept -> int {
-          NV_IF_ELSE_TARGET(NV_IS_HOST,
-                            (printf("Hello from lambda on host! i = %d\n", i);),
-                            (printf("OOPS! still on the device! i = %d\n", i);))
-          return i + 1;
-        });
-
-    // run the task, wait for it to finish, and get the result
-    auto [i] = ex::sync_wait(std::move(start)).value();
-    printf("All done on the host! result = %d\n", i);
-  }
-  catch (cuda::cuda_error const& e)
-  {
-    std::printf("CUDA error: %s\n", e.what());
-  }
-  catch (std::exception const& e)
-  {
-    std::printf("Exception: %s\n", e.what());
-  }
-  catch (...)
-  {
-    std::printf("Unknown exception\n");
-  }
-}