That's the problem. By framework, I mean all of the scripts, prefabs, mechanics, UI, workflow, environment templates, etc. that I have built up within Unity that support creating content for this game. I first developed this for my previous game, refined and expanded it for my current one, and intend to continue using it in the next one. By content, I mean all of the monsters, items, map data and so on that are specific to this game. At this point, I am mostly in drag-and-drop mode for creating content, except for the occasional small feature that I need to code.
Moving all of that to a new engine would be a huge effort and setback, much more than a few days. To give one concrete example, none of them seem to have a clear equivalent to scriptable objects as far as I can tell, which I use very heavily for a variety of purposes (custom assets in Flax might get part of the way there, but not all of the way, I think). I would have to find an alternative implementation for each specific use case. Most would probably have to be converted to JSON files, so I would have to convert the corresponding classes and my content management code to populate from JSON instead of just referencing a scriptable object in the editor. Then I would have to migrate hundreds of scriptable objects specific to this project into JSON files, some of which would reference other JSON files that in turn reference still other JSON files. Then even once it's working, I now have to deal with editing and referencing a few hundred JSON files, instead of working with that data natively in the inspector like I do now.
So that's one example of what would be involved. I have a number of issues with Unity and this merger is one of them, so I am not opposed to finding alternatives, but the alternatives don't seem to be ready yet, and I don't think uprooting my current project is worth the effort and risk. I am more likely to look at other options for my next project, but even then, there is still a lot that I would need to rewrite to make it work. Right now I have a fully-functional, mostly-easy-to-work-with RPG model built in Unity that I am hesitant to throw away without a good solid alternative, and I don't really need the latest versions of Unity anyway. I can stick with 2020 for as long as it's viable, if I need to.