It is too late to reverse this swap and do a split instead, but that would only have affected already identified observations. I think having Isaria as a synonym of Cordyceps has to be right, since the type species determines application of the name, and it has been shown to belong in Cordyceps. Unfortunately, iNat doesn't really have a good way of dealing with misapplications, so this will cause problems. One possible idea to head off some misidentifications is to put as an English common name for Cordycipitaceae "Isaria sensu lato." It probably won't solve everything, but it might help.