[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/cudax/test/execution/test_trampoline_scheduler.cu

@@ -0,0 +1,97 @@
+//===----------------------------------------------------------------------===//
+//
+// 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 "./common/retry.cuh"
+#include "testing.cuh"
+
+namespace ex = ::cuda::experimental::execution;
+
+#if !_CCCL_DEVICE_COMPILATION()
+
+namespace
+{
+struct try_again
+{};
+
+class fails_alot
+{
+  template <class Receiver>
+  struct operation;
+
+public:
+  using sender_concept = ex::sender_t;
+
+  fails_alot() = default;
+
+  __host__ fails_alot(fails_alot&& other) noexcept
+      : counter_(std::move(other.counter_))
+  {}
+
+  __host__ fails_alot(const fails_alot& other) noexcept
+      : counter_(other.counter_)
+  {}
+
+  template <class Receiver>
+  [[nodiscard]] auto connect(Receiver rcvr) const noexcept -> operation<Receiver>
+  {
+    return operation<Receiver>{static_cast<Receiver&&>(rcvr), --*counter_};
+  }
+
+  template <class...>
+  static _CCCL_CONSTEVAL auto get_completion_signatures() noexcept
+  {
+    return ex::completion_signatures<ex::set_value_t(), ex::set_error_t(try_again)>{};
+  }
+
+private:
+  template <class Receiver>
+  struct operation
+  {
+    void start() & noexcept
+    {
+      if (counter_ == 0)
+      {
+        ex::set_value(static_cast<Receiver&&>(rcvr_));
+      }
+      else
+      {
+        ex::set_error(static_cast<Receiver&&>(rcvr_), try_again{});
+      }
+    }
+
+    Receiver rcvr_;
+    int counter_;
+  };
+
+  std::shared_ptr<int> counter_ = std::make_shared<int>(1'000'000);
+};
+
+// #if defined(REQUIRE_TERMINATE)
+// // For some reason, when compiling with nvc++, the forked process dies with SIGSEGV
+// // but the error code returned from ::wait reports success, so this test fails.
+// TEST_CASE("running deeply recursing algo blows the stack", "[schedulers][trampoline_scheduler]") {
+
+//   auto recurse_deeply = retry(fails_alot{});
+//   REQUIRE_TERMINATE([&] { sync_wait(std::move(recurse_deeply)); });
+// }
+// #endif
+
+TEST_CASE("running deeply recursing algo on trampoline_scheduler doesn't blow the stack",
+          "[scheduler][trampoline_scheduler]")
+{
+  ex::trampoline_scheduler sched;
+  auto recurse_deeply = retry(ex::on(sched, fails_alot{}));
+  ex::sync_wait(std::move(recurse_deeply));
+}
+} // namespace
+
+#endif // !_CCCL_DEVICE_COMPILATION()