Goodness-of-Fit Tests for Type-II Censored Samples via the Malmquist Transformation // Goodness-of-fit tests for an arbitrary user-specified continuous distribution under Type-II right- or left-censoring. Implements the transformation-based method of Lin, Huang and Balakrishnan (2008) doi:10.1109/TR.2008.2005860, which uses a property of order statistics due to Malmquist (1950) to convert an r-out-of-n Type-II censored uniform sample into a complete sample of size r, alongside the earlier transformation of Michael and Schucany (1979) doi:10.1080/00401706.1979.10489813. Also implements the direct (untransformed) censored-sample statistics of Barr and Davidson (1973) doi:10.1080/00401706.1973.10489108 and Pettitt and Stephens (1976) doi:10.1093/biomet/63.2.291, and the modified-statistic maximum-likelihood procedure of Chen and Balakrishnan (1995) for testing composite hypotheses. General background on empirical-distribution-function goodness-of-fit methods follows D'Agostino and Stephens (1986, ISBN:982-0-8247-7487-5).