Version 0.3.1

Changes in this version:

• Protein id to taxid mapfile can be supplied to tango assign, allowing the user to assign taxonomy for diamond output that was searched against a database without taxonomic information
• Protein to length mapfile can be supplied to tango assign, allowing the user to normalize percent identity by alignment length fraction for diamond output in the default blast tabular format
• Added option to skip normalizing percent identity by length using tango assign --nolen
• Added option to do diamond blastp search using tango search -m blastp
• More docstrings
• Fixed issue with translated taxids
• Make input format explicit
• Filter by score when reading dataframe