Performance update

Performance is now around TinyScheme levels (slightly above in some cases, slightly below in others); that's not really saying much, since TinyScheme is quite slow, but it's better than where Vesta used to be. I'll focus on cleaning up everything (but core evaluator & FFI take top priority), as well as performance. Hopefully I'll be able to get better performance in Vesta before I work on Enyo & Ceres.