feat(cccl): integrate missing CCCL directories — python/, ci/, .agent/, docs/, test/

Sparse-checkout from NVIDIA/cccl main branch to complete cccl_upstream:

Added:
- python/cuda_cccl/ (226 files) — Python bindings for device-level algorithms
  Critical for muh toolchain: cuda.compute.reduce_into, scan, radix_sort, etc.
  Includes 204 .py files with full test coverage for all 27 algorithms
- ci/ (163 files) — Build/test infrastructure
  build_cub.sh, test_cub.sh, build_and_test_targets.sh, matrix.yaml
  Directly maps to our [INFRA-CI] and [INFRA-BUILD] items
- .agent/skills/ (7 files) — NVIDIA's own agent skills for CCCL
  cccl-style/SKILL.md, cccl-test/SKILL.md, sass-diff/SKILL.md
- docs/ (491 files) — Official CCCL documentation
  CI references, CMake guides, Python compute docs, libcudacxx PTX docs
- test/ (12 files) — Top-level integration tests (cuda_smoke, stdpar)
- Root configs: .clang-format, .clang-tidy, CONTRIBUTING.md, pyproject.toml
- CLAUDE.md symlink → AGENTS.md (NVIDIA's standard)

cccl_upstream now mirrors full NVIDIA/cccl structure:
  Before: 42M (cub + thrust + libcudacxx + cudax + c + examples + benchmarks)
  After:  53M (+python +ci +docs +.agent +test +configs)

This completes the CCCL base needed for:
- [muh-bench] items: ci/util/build_and_test_targets.sh for targeted builds
- [CCCL-verify] items: python/cuda_cccl/tests/ as reference implementations
- [CCCL-test] items: ci/test_cub.sh, ci/test_thrust.sh
- Agent workflow: .agent/skills/ for consistent style and test patterns

cccl_upstream/python/cuda_cccl/cuda/cccl/headers/__init__.py

@@ -0,0 +1,7 @@
+# Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
+#
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+from .include_paths import get_include_paths
+
+__all__ = ["__version__", "get_include_paths"]

cccl_upstream/python/cuda_cccl/cuda/cccl/headers/include/__init__.py

@@ -0,0 +1 @@
+# Intentionally empty

cccl_upstream/python/cuda_cccl/cuda/cccl/headers/include_paths.py

@@ -0,0 +1,51 @@
+# Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
+#
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+import sys
+from dataclasses import dataclass
+from functools import lru_cache
+from importlib.resources import as_file, files
+from pathlib import Path
+from typing import Optional
+
+# type: ignore[import-not-found]
+from cuda.pathfinder import find_nvidia_header_directory
+
+
+@dataclass
+class IncludePaths:
+    cuda: Optional[Path]
+    libcudacxx: Optional[Path]
+    cub: Optional[Path]
+    thrust: Optional[Path]
+
+    def as_tuple(self):
+        # Note: higher-level ... lower-level order:
+        return (self.thrust, self.cub, self.libcudacxx, self.cuda)
+
+
+@lru_cache()
+def get_include_paths(probe_file: str = "cub/version.cuh") -> IncludePaths:
+    cuda_incl = find_nvidia_header_directory("cudart")
+    if cuda_incl is None:
+        raise RuntimeError("Unable to locate CUDA include directory.")
+
+    with as_file(files("cuda.cccl.headers.include")) as f:
+        cccl_incl = Path(f)
+
+    probe_file_path = Path(probe_file)
+    if not (cccl_incl / probe_file_path).exists():
+        for sp in sys.path:
+            cccl_incl = Path(sp).resolve() / "cuda" / "cccl" / "headers" / "include"
+            if (cccl_incl / probe_file_path).exists():
+                break
+        else:
+            raise RuntimeError("Unable to locate CCCL include directory.")
+
+    return IncludePaths(
+        cuda=cuda_incl,
+        libcudacxx=cccl_incl,
+        cub=cccl_incl,
+        thrust=cccl_incl,
+    )