Benchmarks for all models listed in a spreadsheet

[kowaalczyk]

Gather benchmarks for all models into a spreadsheet (model (pos/ner/dep), paper referece, dataset (preferably with download link), used metric(s), metric values achieved in the paper)

[MateuszOlko]

Referring to the notes from our last meeting I assign to this task also @Gizzio as he is responsible (according to the notes) for benchmark for POS-tagger.

Spreadsheet link

[kowaalczyk]

@Gizzio finish adding POS benchmarks here, close the issue after that
Update: we need to NERs from PolEval, POS model @tomekkorbak suggested to check out
Make sure we're able to run models and measure their speed