modelimpact 1.1.0

New features

New ranking-based views: cumulative_gains(), lift_curve(), marginal_profit(), break_even() and impact_summary().
New threshold / classification views: confusion_payoff(), roc_pr() and payoff_grid().
New comparison and robustness tools: compare_models(), bootstrap_profit() and tornado().
All result objects now have ggplot2::autoplot() methods, plus matching plot_*() helper functions, for one-line visualisation.
New run_app() launches an interactive shiny application to explore the package and upload your own data (comma- or semicolon-separated CSV).
profit(), cost_revenue() and roi() gained a positive argument so the event class no longer has to be labelled "Yes", and a prob_accept argument that reconciles them with profit_thresholds().

profit_thresholds() now correctly applies the fp_val (false-positive value) argument, which was previously ignored.