idea: split out budo's guts

[yoshuawuyts]

So I've been working on a bunch of streaming providers for {html,css,js}, and I was wondering what's missing before it could (not saying it should) be used as the heart of budo?

I'm currently including tiny-lr for livereloading out of the box (choojs/bankai#3), and the only difference with budo would be that the glob LR params can't be changed. Would that matter?

Because bankai uses streams everywhere this would also help further #60, which I still think would be pretty great if we could pull off. Probably also simplifies budo's internals a bit, which might be nice?

Lmk what you think / how bankai can improve 😁Thanks!

[mattdesl]

Yeah, I have split some features out already (see watchify-middleware), and ideally I'd like to split out a few more things.

It would be hard to use a module that isn't somewhat purpose-built for budo. There are a lot of subtle things that have grown into budo over time; things like manually handling a debounce for faster incremental bundles, reporting installed watchify version, handling chokidar edge cases, handling babelify and ansi-colored build errors, etc.