Initial Basis (tracking issue)

All of the named unicode charicter classes should be bound to variables in an initial basis, and common expressions should be given sensible names. For example:

`[0-9]` => digit_ascii
`[0-9]+` => number_ascii
`\b\w+\b` => word

The definitions of this initial basis should be displayed prominently in the documentation (It may be worth writing a little script to gen both the docs and the code just to make sure that they never fall out of sync.).
Once function support lands, we should add common functions like `sep(e, e) -> e`, `term(e, e) -> e`, `comma_sep(e)`, `semi_term(e)`, and so on. Until the infastructure is in place for that, new ideas for the initial basis should be added to this issue as a comment.