Joeynmt

- upgrade to sacrebleu 2.0, python 3.7, torch 1.8
- bug fixes:
- heaps in checkpoint maintenance 153
- beam search stopping criterion 149
- removing final BPE merge markers in hypotheses (dsfsi/masakhane-web33)
- keeping best and last ckpts 136
- using utf encoding when opening files 150
- f-style formatting

You can now retrieve the n-best outputs during inference (rather than just the one best translation) and track the latest checkpoint (for continuing training). We also added a colab for training a small translation model on the Tatoeba task. Now operating on Torch v1.8.0 and using deprecated Torchtext dataset implementations from v0.9.

Additions:
- Multi-GPU processing
- Data loading improvements
- Tokenizer integration
- Japanese benchmarks

Stable recurrent and Transformer models. Minor changes and refactoring might happen before v1.0.