Add expected model size controls for spike-and-slab priors

[leninworld]

This PR fixes issue #1177.

Summary

• add expected_num_nonzero and pi_concentration as an explicit convenience pair for spike-and-slab priors
• derive the Beta inclusion-probability shapes from expected model size while preserving existing pi_alpha/pi_beta behavior
• validate incomplete, conflicting, non-finite, and out-of-range configurations
• cover instance-level, call-level, split-precedence, and InstrumentalVariable integration paths

Closes #1177.

Validation

• focused variable-selection and instrumental-variable tests — 47 passed
• commit hooks: licensing, Ruff lint/format, codespell, numpydoc, mypy, and repository checks — passed
• broad suite — 1,249 passed, 5 skipped; 4 failures and 10 errors are isolated to unrelated PyMC time-series validators under Python 3.14
• doctest run — 36 passed before an unrelated Python 3.14/PyMC multiprocessing segfault in panel regression

Validation evidence

Compatibility and scope

Existing defaults remain pi_alpha=2 and pi_beta=2. The convenience parameters are optional, must be supplied together, and cannot be mixed with explicit Beta shape parameters. The expected size controls the prior inclusion probability, not a hard selected-variable count or posterior guarantee.

[github-actions]

👋 Welcome to CausalPy, @leninworld!

Thank you for opening your first pull request! We're excited to have you contribute to the project. 🎉

Here are a few tips to help your PR get merged smoothly:

• ✅ Make sure all CI checks pass (tests, linting, type checking)
• 📝 Run prek run --all-files locally before pushing
• 📖 Check our Contributing Guide for more details

A maintainer will review your changes soon. Thanks for helping make CausalPy better! 🚀

---

💼 LinkedIn Shoutout: Once your PR is merged, we'd love to give you a shoutout on LinkedIn to thank you for your contribution! If you're interested, just drop your LinkedIn profile URL in a comment below.