Deepbgc

- You can now detect BGCs in a *protein* FASTA sequence using `deepbgc pipeline --protein`. This is done by creating an artificial sequence where all proteins are found right after another on the + strand
- You can now prepare *protein* FASTA sequences into a Pfam TSV file using `deepbgc prepare --protein`.
- Fixed ROC and Precision-Recall plots

- add `--prodigal-meta-mode` argument to `deepbgc prepare` and `deepbgc pipeline` to enable detecting genes in short contigs
- fix Python 2.7 incompatibility error when detecting Pfam domains

- Added support for gzipped input sequence files (should end with .gz extension)
- Improved error message when parsing unknown format

- Require Tensorflow < 2.0.0

- Removed forgotten command that printed out all pfam descriptions to STDOUT