XML should be added on monophyly instead of a strict clock

[richelbilderbeek]

Feedback bjoelle:

mrca_prior_to_xml_tracelog, mrca_prior_to_xml_operators, mrca_prior_to_xml_prior_distr, mrca_prior_to_xml_state: these functions only add the parts relative to the strict clock if the MRCA prior is monophyletic, whereas in BEAUTi these parts are added if the MRCA prior has a distribution associated with it, whether it is monophyletic or not.