Add support for Slovenian language

[ajdapretnar]

Proposed enhancements

This issue is to keep track of Slovenian language technologies to support.

• lemmatization, potentially IJS's LemmaGen
• POS tagging, potentially Obeliks
• stopword list (can be provided personally)
• corpora Gigafida, Kres

Then add Orange to MK's list of tools for Slovene language.

[ajdapretnar]

Other suggested tools:

• UDPipe: figure out the documentation and how to use it in Python, I could not
• ReLDI: potentially problematic if one can use it only with API keys (I had issues accessing their service)

[ajdapretnar]

For stopwords, see nltk/nltk_data#54 (comment).

[ajdapretnar]

For sentiment analysis, see CLARIN resources.

[ajdapretnar]

We have added support for Slovenian in Preprocess Text (UDPipe Lemmatizer), Sentiment Analysis, and recently in #504 for Document Embedding. For now, this concludes our quest for supporting Slovene. Other suggestions should be addressed in separate issues.