I wouldn't be sure this is a long term solution. Maintaining a this into perpetuity would probably take a lot of resources that Vivaldi team might not have.
For all possible MV2 extensions this is unsustainable for a small company. Supporting just uBO used by tens of millions of people? pretty doable as it only relies on few hooks into C functions that arent going anywhere.
Unless upstream introduces breaking changes I don't see why not. Google was killing this functionality for policy reasons, not because changes to the renderer required it or something. And LLM analysis is going to make maintaining forks easier going forward.