Add CodSpeed performance benchmarks and CI workflow

[codspeed-hq]

Summary

This PR integrates CodSpeed for continuous performance monitoring of spaghettree's core algorithms.

Changes

• Benchmark suite (tests/test_benchmarks.py): 12 benchmarks covering the performance-critical paths:
  • get_dwm (directed weighted modularity) at small, medium, and large matrix scales
  • AdjMat.from_call_tree construction
  • Merge pair detection (get_merge_pairs, get_top_suggested_merges)
  • Full community optimisation pipeline (optimise_communities)
  • CST parsing pipeline: entity extraction, non-native call filtering, and call tree creation
• CI workflow (.github/workflows/codspeed.yml): Runs benchmarks in simulation mode on pushes to main, pull requests, and manual dispatch. Uses OIDC authentication.
• pytest-codspeed added as a dev dependency.
• CodSpeed badge added to the README.

Notes

• The repo-mapper-rs dev dependency fails to build in current environments (broken maturin config in the published package). The CI workflow and benchmark runs exclude it via --no-install-package repo-mapper-rs since it is not needed for benchmarking.
• Benchmarks override the default pytest addopts (coverage settings) to avoid failing on coverage thresholds, which are not relevant for performance runs.

Next steps

• Once merged, CodSpeed will automatically track performance on every PR and push to main.
• Visit the CodSpeed dashboard to view flamegraphs and performance trends.
• Consider adding benchmarks for the full optimise_entity_positions pipeline as the project grows.

[codspeed-hq]

Congrats! CodSpeed is installed 🎉

🆕 12 new benchmarks were detected.

You will start to see performance impacts in the reports once the benchmarks are run from your default branch.

Detected benchmarks

• test_bench_extract_entities: 18.8 ms
• test_bench_get_dwm_large_random: 254 µs
• test_bench_adjmat_from_call_tree_medium: 298.6 µs
• test_bench_adjmat_from_call_tree_small: 247.4 µs
• test_bench_get_dwm_medium: 147.9 µs
• test_bench_filter_non_native_calls: 6 ms
• test_bench_create_call_tree: 152.9 µs
• test_bench_get_top_suggested_merges: 3.6 ms
• test_bench_get_merge_pairs: 5.3 ms
• test_bench_get_dwm_small: 140.8 µs
• test_bench_optimise_communities_small: 1.7 ms
• test_bench_optimise_communities_medium: 9.7 ms

---