Unify pp_token of Text_tokenizer with other pp

[NeoKaios]

The old pp_token of Text_tokenizer didn't have a prime even though it was a pretty printer for token with_loc

[nberth]

I think the only change that is actually needed it the addition or removal of primes for pp_token functions. The rest makes things just a little bit too messy.

[nberth]

Still not fan of the change to the type names. In the context of the scanner/parser, type token = … with_loc was fine. The main issue was only in the name of pretty-printing functions that are exported by the library. But ok anyways.

[nberth]

Thanks!