Execution modes

single generates files, runs OPM Flow, creates benchmark data, and optionally creates figures. files creates inputs only and cannot use -f all. data processes an existing simulation. ert and everest run their corresponding history-matching workflows. fair uses fixed benchmark comparison settings and rejects conflicting input, figure, time, and experiment options. none skips simulation work and is useful for plotting existing results.