setting start char index in lexer rules

Seems unnecessary as Lexer.tokenStartChar or whatever is available. Are we setting start unnecessarily i lexer rules?

Oh, maybe we need in case they call a lexer rule and ask for token return value.