Flatten plot API: replace Plotter classes with module-level functions

[Jammy2211]

Summary

Replace the class-based plotting API (NestPlotter, MCMCPlotter, MLEPlotter, SamplesPlotter) with simple module-level functions. Users now call aplt.corner_anesthetic(samples=samples) instead of instantiating a plotter object. All cross-cutting concerns (test-mode skipping, convergence error handling) are preserved as decorators in a new plot_util.py.

API Changes

The autofit.plot module no longer exports Plotter classes. All plotting is now done via flat functions that take samples as the first argument, with optional path, filename, and format kwargs for output control.

• Removed: NestPlotter, MCMCPlotter, MLEPlotter, SamplesPlotter, Output
• Added: corner_cornerpy(), corner_anesthetic(), subplot_parameters(), log_likelihood_vs_iteration(), output_figure()

See full details below.

Test Plan

• pytest test_autofit/ -x — 390 passed (1 pre-existing failure in test_covariance.py also fails on main_build)
• All modified files pass py_compile syntax check
• grep confirms no stale references to old Plotter classes in source

📋 Full API Changes (for automation & release notes)

Removed

• aplt.NestPlotter class — use aplt.corner_anesthetic() function instead
• aplt.MCMCPlotter class — use aplt.corner_cornerpy() function instead
• aplt.MLEPlotter class — use aplt.subplot_parameters() / aplt.log_likelihood_vs_iteration() instead
• aplt.SamplesPlotter class — base class, no longer needed
• aplt.Output class — output now handled via path, filename, format kwargs on each function
• autofit.non_linear.plot.output module — deleted
• autofit.non_linear.plot.mcmc_plotters module — deleted

Added

• aplt.corner_cornerpy(samples, path=None, filename="corner", format="show", **kwargs) — corner plot via corner.py
• aplt.corner_anesthetic(samples, path=None, filename="corner_anesthetic", format="show", **kwargs) — corner plot via anesthetic
• aplt.subplot_parameters(samples, use_log_y=False, use_last_50_percent=False, path=None, filename="subplot_parameters", format="show") — parameter vs iteration subplots
• aplt.log_likelihood_vs_iteration(samples, use_log_y=False, use_last_50_percent=False, path=None, filename="log_likelihood_vs_iteration", format="show") — log likelihood vs iteration
• aplt.output_figure(path=None, filename="figure", format="show") — utility for saving/showing figures
• autofit.non_linear.plot.plot_util module — shared decorators (skip_in_test_mode, log_plot_exception) and output_figure

Migration

• Before: plotter = aplt.NestPlotter(samples=s); plotter.corner_anesthetic()
• After: aplt.corner_anesthetic(samples=s)
• Before: plotter = aplt.MCMCPlotter(samples=s); plotter.corner_cornerpy(**kw)
• After: aplt.corner_cornerpy(samples=s, **kw)
• Before: plotter = aplt.MLEPlotter(samples=s); plotter.subplot_parameters(use_log_y=True)
• After: aplt.subplot_parameters(samples=s, use_log_y=True)
• For file output, replace Output(path=p, format="png") with kwargs: path=p, format="png" on each function call

🤖 Generated with Claude Code