I do not think you can build something exactly equivalent to a drivechain using BitVM. We came up with a few ideas for a form of SPV sidechain. I'm not confident that any of them work, but I think we are close. With a few refinements on our initial ideas, we very well might get something workable.

An SPV sidechain is a kind originally proposed in this 2015 whitepaper: https://blockstream.com/sidechains.pdf

They were thought to require a soft fork. Maybe not anymore. In comparison to a drivechain, an SPV drivechain is probably slightly worse. SPV sidechains rely on an assumption that underpins both bitcoin's doublespend protection and bip300's hashrate escrow: the assumption that 51% of miners will not collude to steal.

However, an SPV sidechain, this assumption is implicit (not explicit) and it does not provide the 3 month buffer that drivechain provides. Miners can "quickly" steal from an SPV sidechain but the whitepaper ignores this because it assumes miners will not do that.

To me there is a sense in which that makes an SPV sidechain slightly worse that a drivechain. Still, I'll happily take what I can get. If we can think of a way to do SPV sidechains via BitVM, I'll get for it, but I'll still want drivechains because I think they are still a slight improvement. Maybe like a 4% improvement.