Seol mar théacs é seo: Bayesian auxiliary variable models for binary and multinomial regression