Text this: Classical Versus Bayesian Error-Controlled Sampling Under Lognormal Distributions with Type II Censoring