fn scan_string_token_len(bytes: &[u8], start: usize) -> usize
Length of a string token. May be a normal '...' or extended E'...' / e'...' form (the E prefix is part of the token offset).
'...'
E'...'
e'...'