perf: cache call graph call nodes

[mldangelo-oai]

Summary

• cache repeated _iter_call_nodes() AST walks by function node
• include the new cache in the existing source-sensitive invalidation sweep
• add a regression that proves the second lookup reuses the cached walk

Benchmarks

• repeated same-function call-node walk, 5,000 lookups x 5 passes:
  • before: 1.000113s median
  • after: 0.000891s median
• matched cProfile run on tests/assets/exploits/exploit_ultimate_50pct.pkl from origin/main:
  • before: 82.470s
  • after: 66.972s

Validation

• UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run pytest packages/modelaudit-picklescan/tests/test_call_graph_import_statements.py::test_iter_call_nodes_reuses_cached_walk -q
• UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run pytest packages/modelaudit-picklescan/tests/test_call_graph_import_statements.py -q
• UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run ruff format packages/modelaudit-picklescan/src packages/modelaudit-picklescan/tests
• UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run ruff check --fix packages/modelaudit-picklescan/src packages/modelaudit-picklescan/tests
• UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run mypy modelaudit/ packages/modelaudit-picklescan/src packages/modelaudit-picklescan/tests tests/ (hits existing mypy 1.20.0 internal error)
• PROMPTFOO_DISABLE_TELEMETRY=1 UV_CACHE_DIR=/tmp/modelaudit-uv-cache uv run pytest -n auto -m "not slow and not integration" --maxfail=1 (reproduces the existing timing-sensitive directory-scan and validation-performance sentinels in this environment)

[github-actions]

Workflow run and artifacts

Performance Benchmarks

Compared 19 shared benchmarks with a regression threshold of 15%.
Status: 0 regressions, 0 improved, 19 stable, 0 new, 0 missing.
Aggregate shared-benchmark median: 359.73ms -> 361.96ms (+0.6%).

Benchmark	Target	Size	Files	Baseline	Current	Change	Status
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_dangerous_global_payloads[malicious_reduce]	malicious_reduce	52 B	1	376.2us	399.2us	+6.1%	stable
tests/benchmarks/test_scan_benchmarks.py::test_validate_file_type_pytorch_zip	state_dict.pt	1.5 MiB	1	52.5us	55.3us	+5.4%	stable
tests/benchmarks/test_scan_benchmarks.py::test_skip_filter_plain_text_files	-	4.6 KiB	256	13.16ms	13.80ms	+4.9%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_nested_payloads[nested_raw]	nested_raw	78 B	1	122.5us	126.4us	+3.2%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_multi_stream_padded_payload	multi_stream_padded	4.1 KiB	1	448.5us	462.8us	+3.2%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_chunked_stream	chunked_stream	278.2 KiB	1	18.37ms	17.88ms	-2.7%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_dangerous_global_payloads[stack_global]	stack_global	21 B	1	322.5us	330.8us	+2.6%	stable
tests/benchmarks/test_scan_benchmarks.py::test_scan_pytorch_zip	state_dict.pt	1.5 MiB	1	30.14ms	30.79ms	+2.2%	stable
tests/benchmarks/test_scan_benchmarks.py::test_scan_safe_pickle	safe_model.pkl	49.4 KiB	1	11.35ms	11.11ms	-2.1%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_nested_payloads[nested_base64]	nested_base64	98 B	1	129.9us	132.2us	+1.8%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_safe_payloads[safe_large]	safe_large	278.2 KiB	1	16.12ms	15.86ms	-1.6%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_opcode_budget_tail_payload	opcode_budget_tail	14 B	1	91.1us	92.3us	+1.3%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_safe_payloads[safe_small]	safe_small	68 B	1	77.4us	76.5us	-1.2%	stable
tests/benchmarks/test_scan_benchmarks.py::test_detect_file_format_safe_pickle	safe_model.pkl	49.4 KiB	1	31.7us	31.3us	-1.1%	stable
tests/benchmarks/test_scan_benchmarks.py::test_scan_duplicate_directory	duplicate-corpus	840.0 KiB	81	193.64ms	195.45ms	+0.9%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_hidden_suspicious_string_budget	hidden_suspicious_string	8.0 KiB	1	620.2us	625.0us	+0.8%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_nested_payloads[nested_hex]	nested_hex	130 B	1	139.0us	139.9us	+0.6%	stable
tests/benchmarks/test_picklescan_benchmarks.py::test_picklescan_safe_payloads[long_benign_string]	long_benign_string	1.0 MiB	1	1.06ms	1.06ms	-0.2%	stable
tests/benchmarks/test_scan_benchmarks.py::test_scan_mixed_directory	mixed-corpus	1.7 MiB	54	73.49ms	73.53ms	+0.1%	stable