I would love to know how much of the budget (aka time) for a game that goes to research, feature design, proof of concepts etc, and how much that goes into actually programming features, graphical design etc. in something resembling the final state. I doubt we will ever get to know that, and it's probably not entirely clear for Paradox either since they will overlap, but I'm pretty sure porting a feature is generally significantly less labour intensive than creating the same feature from scratch. Be it the map, events or unit models with all the research that can go into it, or a basic mechanic such as moving a unit from one province to another with all the restrictions and attributes that comes with it.
It is not unreasonable to expect a significant portion of features added through DLCs to be reintroduced in the base game in the sequel. How big that portion should be probably depends on which features people like. I would for instance be perfectly happy to see mission trees, which is a very significant portion of the content added over the past five years, completely gone from EU5.