Clean up dependencies for useless (in our cases) file parsers
Tika is shipped with almost all its parsers and dependencies, however a lot of them are not related to the kind of files we want to be able to handle as part of the WebLab, or that cannot be handled yet in the way the process is done. For instance, there are Compression and packaging formats (while we generate a single document and not a composed resource for the moment), Feed and Syndication formats, Java class files and archives, Source code, CAD formats, Font formats, Scientific formats, Executable programs and libraries, Database formats.
On top of that the newest version are trying to integrate other features like Natural Language Processing or Image and Video object recognition inside Tika package.
We should be able to reduce the size of the web service package (and save a few time each time we upload/download/unzip this package and the bundle