dsBinVal: Conducting distributed ROC analysis using DataSHIELD

Methods to conduct distributed ROC and calibration analyses. The basis is the DataSHIELD <https://www.datashield.org> infrastructure for distributed computing. This package provides the calculation of the ROC-GLM <https://www.jstor.org/stable/2676973?seq=1> as well as AUC confidence intervals <https://www.jstor.org/stable/2531595?seq=1>. To assess the calibration, methods to calculate the Brier score and calibration curves are part of the package. The last part of the package are methods to push models and predict models at the DataSHIELD server which is necessary for the analysis. DataSHIELD uses the option `datashield.privacyLevel` to indicate the minimum amount of numbers required to be allowed to share an aggregated value of these numbers. Instead of setting the option, we directly retrieve the privacy level from the DESCRIPTION <https://github.com/difuture-lmu/dsBinVal/blob/master/DESCRIPTION> file each time a function calls for it. This option is set to 5 by default. Methodological details can be viewed in Schalk et al. (2022) <arXiv:2203.10828>.