Practical approaches to Bayesian sample size determination in non‐inferiority trials with binary outcomes