Conventions for representing clinical trials supporting and Edge

[mbrush]

As noted in the now closed issue #398 (and the current supporting_publications_specification) - we are currently allowing use of the publications edge property to capture CURIEs for clinical trials that support a given edge. But there was consensus that we will eventually want a separate dedicated property for referencing supporting trails (something like supporting_studies) .

Details/discussion to date can be found in comments on #398 which is now closed - so all further discussion can happen here. We will address this post Sept release, and update the spec and Biolink models accordingly.