Let's deprecate `mithril-stream`

[dead-claudia]

Currently, mithril-stream is almost never updated. I just got publish rights a few days ago for that and ospec and have since updated them both.

I suggest we deprecate mithril-stream when v2 goes stable since we've already been recommending people to instead use mithril/stream directly from mithril itself and 99% of users are also already using Mithril itself. Our docs don't even mention mithril-stream anymore mod a small docs bug in the stream docs that I've since fixed in the repo.

Alternatively, we could tell people to start using mithril-stream and drop mithril/stream from the mithril package starting v2, to push people over to that out-of-core package.

---

BTW, if you're confused, I'm specifically referring to this npm package, not the module itself.

[spacejack]

I think I prefer the option to drop mithril-stream. mithril/stream is optional as it is.

One point in favour of continuing mithril-stream might be if it is useful as a Fantasy Land compliant stream lib on its own. One use-case might be if it is the smallest FL streams lib available? Are there any other reasons to use it over flyd or other alternatives?

[dead-claudia]

@spacejack Ours is pretty barebones - large enough to be useful but small enough it doesn't add significant weight. This is useful in the front end if you just want something good enough without being excessive, but code size isn't nearly as much of an issue in the backend, so I'd say Mithril probably isn't offering anything Flyd isn't for server-side.