Fix bug with base NLLB translation

[TaperChipmunk32]

This PR fixes issue #778, allowing the base NLLB model to translate with the correct language codes.

src_lang and trg_lang are added as arguments for _create_inference_model and _configure model in silnlp.nmt.hugging_face_config.py to set the language codes properly in _configure_model. The test language codes are now used instead of the validation language codes, unless a different code is provided using the --trg-iso or --src-iso options for silnlp.nmt.translate.py

---

This change is 

[ddaspit]

Reviewed 1 of 1 files at r1, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @benjaminking)

---

silnlp/nmt/hugging_face_config.py line 909 at r1 (raw file):

        # Change specific variables based on the type of model
        model, tokenizer = self._configure_model(
            model, tokenizer, self._config.test_src_lang, self._config.test_trg_lang

During training, we want to use the validation language.

[ddaspit]

Reviewed 1 of 1 files at r2, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @benjaminking)