Example datasets

[kasnerz]

Prepare a representative set of example datasets released along with the package.

Currently, we have:

• the dummy dataset, which is sort of a MWE,
• five data-to-text datasets from our quintd experiments.
• logicnlg #28 #24 #21

For the release, we want to have:

• 1-2 data-to-text datasets showcasing the possibilities of data visualization (→ filtering the existing ones should be enough),
• 2-3 datasets from other domains: speech-to-text, fact checking, machine translation, etc.,
• a more representative version of the dummy dataset (with a more instructive content than just lorem ipsum).