Default prior run as part of `lalinference_pipe`
It'd be good to have the pipeline generate a set of prior samples corresponding to the prior that's used for any given run, and then produce a comparison page with the results.
If using the ROQ bases an alternative could be to generate the prior files once, and then just read them in (since a non-rescaled ROQ run will have a known prior). It might be simpler to just redo the prior every time however...
Ping @charlie.hoy for implementation into PEsummary (in connection to !668 (merged))
Also @vivien