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/ci/util/python/common_arg_parser.sh

@@ -0,0 +1,58 @@
+# Argument parser for Python CI scripts.
+parse_python_args() {
+    # Initialize variables
+    py_version=""
+    # ctk_mode carries the -ctk-mode value; empty means the default ("pinned").
+    ctk_mode=""
+
+    while [[ $# -gt 0 ]]; do
+        case $1 in
+            -py-version=*)
+                py_version="${1#*=}"
+                shift
+                ;;
+            -py-version)
+                if [[ $# -lt 2 ]]; then
+                    echo "Error: -py-version requires a value" >&2
+                    return 1
+                fi
+                py_version="$2"
+                shift 2
+                ;;
+            -ctk-mode=*)
+                ctk_mode="${1#*=}"
+                # Reject an explicit-but-empty value (e.g. `-ctk-mode=`): a lane
+                # that wants the default omits the flag entirely, so an empty
+                # value signals a malformed generated argument -- fail loudly.
+                if [[ -z "${ctk_mode}" ]]; then
+                    echo "Error: -ctk-mode requires a value" >&2
+                    return 1
+                fi
+                shift
+                ;;
+            -ctk-mode)
+                if [[ $# -lt 2 || -z "$2" ]]; then
+                    echo "Error: -ctk-mode requires a value" >&2
+                    return 1
+                fi
+                ctk_mode="$2"
+                shift 2
+                ;;
+            *)
+                # Unknown argument, ignore
+                shift
+                ;;
+        esac
+    done
+
+    # Export for use by the calling script (py_version and ctk_mode are its inputs).
+    export py_version ctk_mode
+}
+
+require_py_version() {
+    if [[ -z "$py_version" ]]; then
+        echo "Error: -py-version is required" >&2
+        [[ -n "$1" ]] && echo "$1" >&2
+        return 1
+    fi
+}