Follow fancygam to its new name, which had quietly disabled it

[chross22]

fancygam was renamed fancyfx when it grew past GAMs. GitHub still redirects chross22/fancygam, so Remotes: chross22/fancygam kept installing something — but what it installs declares Package: fancyfx, so the requireNamespace("fancygam") behind has_fancygam() has returned FALSE on every fresh install since the rename.

That gate guards a diagnostic rather than the run, so nothing failed. GAM smooth plots simply stopped being written and gam_smooths.png stopped appearing, with no error and no message. It looks fine on a machine that still has the old fancygam 0.0.0.9000 sitting in its library next to the new package — which is exactly why it went unnoticed.

The change

Mechanical: fancygam → fancyfx across DESCRIPTION, the two call sites, the app, the tests and the README. The one call that matters needed no change — combinePlots() keeps its first three arguments and has only gained optional ones. It gained interval, defaulting to "auto", so the smooths now come with their standard error bands.

model_engine_fit()'s example moves from plotSmooths() to plotEffects(). plotSmooths() still works and produces an identical plot, but it is deprecated now and warns at runtime, and a documented example is the wrong place to teach the old name.

Verification

Verified by drawing the plot, not by loading the namespace: a GAM fitted on the mock config writes a three-panel figure with a curve, an error band and a data rug per term. Checked against fancyfx 0.3.0, the current release. Full suite passing.

Not in this PR

fancyfx 0.3.0 also added plotROC(), plotThreshold(), plotImportance() and permutation_importance(), which overlap taupatch's own diagnostics. They take (model, newdata), whereas taupatch computes all of these from pooled held-out cross-validation predictions and documents that heavily. Adopting them with newdata = training data would quietly make every diagnostic flattering. They do have a folds argument, so it may be reconcilable — worth investigating deliberately rather than folding into a rename.

🤖 Generated with Claude Code