Dataset setup script

[dirkweissenborn]

Our current dataset setup is a bit cumbersome in some cases.

We need a CLI that:

1. shows the user which dataset are available for which tasks
2. allows for automagical setup of the datasets, such that the user can use them right away with the training script ; this entails that the script downloads the data and converts it to jack format if necessary (e.g., it is not necessary for squad and snli-like data).