Use hyphenization algorithm.

Pursuant to the discussion in #149, I have started work on incorporating [Hypher](https://github.com/bramstein/hypher) into jetzt. I've made an initial checkin on my [dev branch](https://github.com/alaska/jetzt/tree/dev) that seems to work pretty well, but it's currently hard-coded for en-us. Should I support any other languages, or should I just push on with this as-is, making sure I fall back to the default behavior in case of other languages?

Read the discussion in #149 and inspect the initial Hypher work on the linked dev branch. Resolve whether additional languages are required and how unsupported languages should fall back, then confirm the chosen behavior works beyond the current en-us case.

Written by the indexing model from the issue text.