[CCCL] Add missing CCCL components: c2h, nvbench_helper, cmake, cudax, AGENTS.md

Added 863 files from NVIDIA/cccl sparse checkout:
- c2h/ (27 files): Catch2 test helpers — generators, validators, runner
- nvbench_helper/ (10 files): Benchmark harness utilities
- cmake/ (29 files): CMake presets and build helpers
- cudax/ (794 files): Experimental CUDA extensions
- AGENTS.md: NVIDIA's official AI agent instructions for CCCL
- CMakePresets.json: Standardized build configurations
- cccl-version.json: Version tracking

Also added CCCL_ASSET_MAP.md mapping all 4295 CCCL files to
competition value and PRD items.

cccl_upstream now covers 100% of competition-critical assets:
- 27 tuning headers (SM80/90/100 benchmark data)
- 32 dispatch headers (algorithm implementations)
- 60 Thrust examples (correctness verification)
- 217 CUB Catch2 tests (regression matrix)
- 153 CUB benchmarks (parameter space search)
- 18 CUB examples (API verification)
- 27 test helpers + benchmark harness
- 794 cudax experimental extensions

cccl_upstream/cudax/include/cuda/experimental/__detail/type_traits.cuh

@@ -0,0 +1,119 @@
+//===----------------------------------------------------------------------===//
+//
+// 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) 2025 NVIDIA CORPORATION & AFFILIATES.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef __CUDAX_DETAIL_TYPE_TRAITS_CUH
+#define __CUDAX_DETAIL_TYPE_TRAITS_CUH
+
+#include <cuda/std/detail/__config>
+
+#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
+#  pragma GCC system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
+#  pragma clang system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
+#  pragma system_header
+#endif // no system header
+
+#include <cuda/std/__cccl/unreachable.h>
+#include <cuda/std/__concepts/concept_macros.h>
+#include <cuda/std/__type_traits/decay.h>
+#include <cuda/std/__type_traits/enable_if.h>
+#include <cuda/std/__type_traits/integral_constant.h>
+#include <cuda/std/__type_traits/is_callable.h>
+#include <cuda/std/__type_traits/is_constructible.h>
+#include <cuda/std/__type_traits/is_copy_constructible.h>
+#include <cuda/std/__type_traits/is_move_constructible.h>
+#include <cuda/std/__type_traits/is_nothrow_constructible.h>
+#include <cuda/std/__type_traits/is_nothrow_copy_constructible.h>
+#include <cuda/std/__type_traits/is_nothrow_move_constructible.h>
+#include <cuda/std/__type_traits/is_same.h>
+#include <cuda/std/__type_traits/is_valid_expansion.h>
+#include <cuda/std/__utility/declval.h>
+
+#include <cuda/std/__cccl/prologue.h>
+
+namespace cuda::experimental
+{
+using ::cuda::std::__declfn_t;
+using ::cuda::std::decay_t;
+
+template <class _Ty, bool _Nothrow = true>
+[[noreturn]] _CCCL_HOST_DEVICE_API auto __declfn() noexcept(_Nothrow) -> _Ty
+{
+  _CCCL_ASSERT(false, "__declfn should never be called at runtime.");
+  _CCCL_UNREACHABLE();
+}
+
+template <class _Ty, class _Uy>
+_CCCL_CONCEPT __same_as = ::cuda::std::_IsSame<_Ty, _Uy>::value;
+
+template <class _Ty, class _Uy>
+_CCCL_CONCEPT __not_same_as = !::cuda::std::_IsSame<_Ty, _Uy>::value;
+
+template <class _Ty, class... _Us>
+_CCCL_CONCEPT __one_of = (__same_as<_Ty, _Us> || ...);
+
+template <class _Ty, class... _Us>
+_CCCL_CONCEPT __none_of = (__not_same_as<_Ty, _Us> && ...);
+
+#if _CCCL_HAS_CONCEPTS()
+
+template <template <class...> class _Fn, class... _Ts>
+_CCCL_CONCEPT __is_instantiable_with = requires { typename _Fn<_Ts...>; };
+
+template <class _Fn, class... _As>
+_CCCL_CONCEPT __callable = requires(__declfn_t<_Fn> __fn, __declfn_t<_As>... __as) { __fn()(__as()...); };
+
+#else // ^^^ _CCCL_HAS_CONCEPTS() ^^^ / vvv !_CCCL_HAS_CONCEPTS() vvv
+
+template <template <class...> class _Fn, class... _Ts>
+_CCCL_CONCEPT __is_instantiable_with = ::cuda::std::_IsValidExpansion<_Fn, _Ts...>::value;
+
+template <class _Fn, class... _As>
+_CCCL_CONCEPT __callable = ::cuda::std::__is_callable_v<_Fn, _As...>;
+
+#endif // !_CCCL_HAS_CONCEPTS()
+
+template <class _Fn, class... _As>
+_CCCL_CONCEPT __constructible = ::cuda::std::is_constructible_v<_Fn, _As...>;
+
+template <class... _As>
+_CCCL_CONCEPT __decay_copyable = (::cuda::std::is_constructible_v<decay_t<_As>, _As> && ...);
+
+template <class... _As>
+_CCCL_CONCEPT __movable = (::cuda::std::is_move_constructible_v<_As> && ...);
+
+template <class... _As>
+_CCCL_CONCEPT __copyable = (::cuda::std::is_copy_constructible_v<_As> && ...);
+
+template <class _Fn, class... _As>
+_CCCL_CONCEPT __nothrow_callable = ::cuda::std::__is_nothrow_callable_v<_Fn, _As...>;
+
+template <class _Ty, class... _As>
+_CCCL_CONCEPT __nothrow_constructible = ::cuda::std::is_nothrow_constructible_v<_Ty, _As...>;
+
+template <class... _As>
+_CCCL_CONCEPT __nothrow_decay_copyable = (::cuda::std::is_nothrow_constructible_v<decay_t<_As>, _As> && ...);
+
+template <class... _As>
+_CCCL_CONCEPT __nothrow_movable = (::cuda::std::is_nothrow_move_constructible_v<_As> && ...);
+
+template <class... _As>
+_CCCL_CONCEPT __nothrow_copyable = (::cuda::std::is_nothrow_copy_constructible_v<_As> && ...);
+
+template <class... _As>
+using __nothrow_decay_copyable_t _CCCL_NODEBUG_ALIAS = ::cuda::std::bool_constant<__nothrow_decay_copyable<_As...>>;
+
+using ::cuda::std::__call_result_t;
+} // namespace cuda::experimental
+
+#include <cuda/std/__cccl/epilogue.h>
+
+#endif // __CUDAX_DETAIL_TYPE_TRAITS_CUH

cccl_upstream/cudax/include/cuda/experimental/__detail/utility.cuh

@@ -0,0 +1,54 @@
+//===----------------------------------------------------------------------===//
+//
+// 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) 2025 NVIDIA CORPORATION & AFFILIATES.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef __CUDAX_DETAIL_UTILITY_H
+#define __CUDAX_DETAIL_UTILITY_H
+
+#include <cuda/std/detail/__config>
+
+#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
+#  pragma GCC system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
+#  pragma clang system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
+#  pragma system_header
+#endif // no system header
+
+#include <cuda/std/__type_traits/is_callable.h>
+#include <cuda/std/__type_traits/type_list.h>
+#include <cuda/std/__utility/declval.h>
+#include <cuda/std/__utility/move.h>
+
+#include <cuda/experimental/__detail/type_traits.cuh>
+
+#include <cuda/std/__cccl/prologue.h>
+
+namespace cuda::experimental
+{
+// NOLINTBEGIN(misc-unused-using-decls)
+using ::cuda::std::declval;
+// NOLINTEND(misc-unused-using-decls)
+
+struct _CCCL_TYPE_VISIBILITY_DEFAULT no_init_t
+{
+  _CCCL_HIDE_FROM_ABI explicit no_init_t() = default;
+};
+
+_CCCL_GLOBAL_CONSTANT no_init_t no_init{};
+
+using uninit_t CCCL_DEPRECATED_BECAUSE("Use cuda::experimental::no_init_t instead") = no_init_t;
+
+// TODO: CCCL_DEPRECATED_BECAUSE("Use cuda::experimental::no_init instead")
+_CCCL_GLOBAL_CONSTANT no_init_t uninit{};
+} // namespace cuda::experimental
+
+#include <cuda/std/__cccl/epilogue.h>
+
+#endif // __CUDAX_DETAIL_UTILITY_H