Local Influence Diagnostics for the Extreme-Value Birnbaum- Saunders Regression Model // Implements local influence diagnostics for the Extreme-Value Birnbaum-Saunders (EVBS) regression model: joint maximum likelihood estimation, conformal normal curvature diagnostics under three perturbation schemes (case-weight, response variable, and explanatory variable), randomized quantile residuals with simulation envelope, Monte Carlo simulation utilities, and publication- quality density and diagnostic plots. The methods are described in Ospina, Lima, Barros, and Macedo (2026, submitted) and are applied to monthly maximum wind gust data from Itajai, Brazil.