Unify Poker.Table with poker-game

[santiweight]

I see no reason, currently, why the two can't be unified. In private discussions we generally agreed that poker-game would supercede Poker.Table. Is this still your feeling @tonyday567 - I hope I'm not misrepresenting!

Some TODOs of potential issues:

• [ ] Pretty representations are different (especially because poker-game carries a little more logic with it (though still the bare minimum for full poker afaik)
• [ ] use lens instead of micro-lens. I don't like this one - I think micro-lens is probably better... I've made an issue over at poker-game because we should also consider optics: https://github.com/santiweight/poker-game/issues/1
• [ ] poker-game does not yet support showdown. This is mostly a design issue (I haven't decided what the API will look like). I expect as a first pass I'll accommodate the poker-fold API since that has been designed by an intelligent human and we can always come back to it if need be.

Some issues to consider:

• Speed. While I think we should think about speed - I think neither Poker.Table nor poker-game are particularly performance conscious (both use [a] for example). We should revisit this once we have a way to judge performance without taking guesses...

[tonyday567]

Yep, agreed.

Happy to go with any of the optics right now - it's so confusing and I don't think I'll have any permanent allegiance to one of the camps until record dot syntax happens. I have a slight preference for optics, but mostly because I haven't used it seriously yet, and need to try.

My pretty instances are just for testing. I'm really liking prettyprinter - much more ergonomic than the usual way I process text.