URI.extract is too strong, not limited to URLs, matched real text.
Same issue was present in language detector.

* Count all URLs in text as 23 characters flat, do not count domain part of usernames

* Add new status text counting logic to web UI

* Bring Toot button in line with counter

Both should use stringz I guess

* Use grapheme_length for character count

* Add `Presenters` group to SimpleCov configuration

* Move validators to app/validators, add to simplecov config

Federate spoiler_text using warning attribute on <content /> instead of a <category term="spoiler" />
Clean up schema file from accidental development migrations