- Implemented stubs: latino_tokenize_multiple and latino_tag_adcmultiple (before they were not working at all as designed)
    - Added option _tabSeparatedTitle when loading ADC from file or string and considerably debugged both functions
* Some latino interfaces bugfixes
* Dropped sentiment analisys placeholders
* Other changes
    - Upgraded Latino to the latest version (873 from SVN)
    - Some .NET projects moved to v3.5 framework due to latino upgrade
    - Improvement of SmoothInterfaces - now nodes id are more constant since generated from UID -> db json definition file will not be totally different every time but only those parts that really changed

- wordification widget
- popravki v DBcontext
- database to orange converter deluje na več tabelah

added sdmaleph widget; updated sdmsegs viewer to use simpler result dictionary and to use DataTable.js

exported sdmsegs webservice to ilp package; moved sdm segs rule viewer from workflows/library.py to the ilp package

fixed a problem that occured if a mysql column contains strings with accents - strings that differ only by accents are treated as equal by MySQL  when doing distinct on that column. added a workaround by doing distinct by looking at the binary values instead. also added pretty print for DBContext.

Conflicts:
	workflows/static/js/search.js
	workflows/templates/index.html

I hope I haven't messed up anything. I tried these branches, but, well, I got lost a bit :)

- premaknjeno iskanje vseh vrednosti iz Database Context v Database to Orange Table

fixed the following: check the connection in mysql connect; pull the foreign keys only for the selected db; added escape symbols for columns in SQL queries; moved the attribute value selection dialog out of DB Context widget, since this should be the job of the learner.

added --updateuid option to export_package, which assigns uuids to models with no uuids (new widgets, options, etc).