[egg author here]
Very nice! You even beat us to merging relational e-matching  Re: performance, our current understanding is that “classic” e-matching will be tough to beat for small, simple patterns (which are most patterns!), but I think with some work you can get relational e-matching to parity there.
 Re: performance, our current understanding is that “classic” e-matching will be tough to beat for small, simple patterns (which are most patterns!), but I think with some work you can get relational e-matching to parity there.
Feel free to join the e-graphs zulip to discuss!