stitched together final pipeline architecture before submission

[harmsm]

• Added some unit testing
• Added code to sanitize and check topiary dataframes
• OTT columns are now strings with format ottNUMBER
• Improved reverse blasting
  • can now reverse blast on NCBI
  • cleaned up how reverse blasts are done -- standard data frame structure
  • updated reverse blast calling code to be simpler and more robust
• now catches more OTT errors and throws user-interpretable errors
• switched warnings.warn calls to stdout prints because jupyter often hides warnings silently
• split concept of reverse blast call and nicknames. users can now specify nicknames
• updated fasta read/load capabilities.