More annotations in the MISC column

[rahonalab]

Similarly to the start_char and end_char information, it would be useful to write down annotations such as NER in the MISC column while calling convert_token_dict() Now the default is start_char/end_char, but maybe it would be better to have this as an option, something like

calling_convert_token_dict(misc)

where misc is a list of annotations such as ner, start_char/end_char or multi_ner

[AngledLuffa]

The next version will have NER tags in the misc field of the conll docs

https://github.com/stanfordnlp/stanza/pull/996