Splits text into word tokens, individual punctuation marks, and underscores.
text
List<String> tokenize(String text) { return _tokenRe.allMatches(text).map((m) => m.group(0)!).toList(); }