feat: moderndive R argument parity (correlation methods, newdata, …)

[ismayc]

First of two PRs closing the remaining R argument-level gaps (functions all already existed). This one covers moderndive:

• get_correlation(method=, na_rm=) — Spearman and Kendall rank correlations (via scipy), in addition to Pearson; na_rm toggles per-pair null dropping.
• get_regression_points(newdata=, ID=) — apply the model to a held-out set (a residual is included when the outcome is present in newdata), and use a column as the row identifier.
• get_regression_table(default_categorical_levels=True) — keep the raw statsmodels factor-level term names instead of the prettified "var: level" form.
• gg_parallel_slopes(alpha=) — point transparency (both engines).

Tests cover scipy-matching for each correlation method, na_rm=False → nan, newdata with/without outcome, ID from a source column and from newdata, all error paths, raw vs prettified terms, and the alpha plots. Docs updated (correlation methods + newdata prediction). 291 tests, 100% coverage, ruff clean, docs -W clean.

The companion infer arg-parity PR (median point null, prop_test options, weighted sampling, generate(variables=), shade fill, dens_color) is next.

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (b4944a0) to head (98d4c8c).

Additional details and impacted files

@@            Coverage Diff            @@
##              main       #18   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           22        22           
  Lines         1492      1546   +54     
=========================================
+ Hits          1492      1546   +54     

Files with missing lines	Coverage Δ
moderndive/correlation.py	100.00% <100.00%> (ø)
moderndive/modeling.py	100.00% <100.00%> (ø)
moderndive/plots/models.py	100.00% <100.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.