Pharokka

------------------

* Fixes bug with pharokka_plotter.py, which would crash if the phage had tmRMAs or CRISPRs.
* Fixes bug where integration & excision fwd strand CDS would not be plotted in the correct colour
* Adds tmRNAs and CRISPRs to pharokka_plotter.py.

------------------

* Adds tRNAs to pharokka_plotter.py.
* Adds the -s split mode option with metagenome mode, this will output separate single fastas, gff and genbank files along with -m. It is ideally used for situations where you have bulk phage isolates you want to annotate in one go.

------------------

* Adds pharokka_plotter.py to create plots with pyCirclize.
* Fixes issue with VFDB and CARD counts in _cds_functions.tsv being 0 even is a virulence factor or AMR gene is detected.
* Adds better error checking for --threads.

------------------

* Minor update to fix Biopython version <=1.80, due to a breaking change with 1.81 and bcbio-gff this [issue](https://github.com/chapmanb/bcbb/issues/136).

------------------

* Adds the functionality of mapping each contig against the INPHARED database using mash (https://github.com/RyanCook94/inphared). The top hit for each contig (under a maximum mash distance threshold of 0.2) is kept.
* New database adding INPHARED.
* Replaced prodigal with pyrodigal as it is being actively maintained and used by bakta.
* Adds --citation.
* Adds checks for dependencies.
* Adds --terminase terminase mode to re-orient a single contig phage to begin with a certain orientation and coordinate (most commonly, the large terminase subunit). With this, you must also specify --terminase_strand the strand of the terL gene and --terminase_start the start coordinate.
* All locus tags end with 4 digits (trailing zeros) in order to play nice with vConTACT2 and start with 1 not 0.
* In meta mode, the locus tags now begin with the contig header, not a random string (or chosen prefix).
* Cleans up the .tbl so it should automatically be accepted by NCBI Bankit.

------------------

* Renames the CDS output files to *.faa for amino acids and *.ffn for nulceotide sequences
* Implementation of consistent CDS name (equal to the locus_tag) across all output files
* Creates terL.faa and terL.ffn, which contain the sequences of any identified terminase large subunit CDSs
* Passes multithreading to PHANOTATE and tRNAscan-SE in meta mode indicated by flag -m, which provides approximately a t-fold improvement in run-time for large metavirome datasets, where t is the number of threads.