Fix typo causing bug in main script

Update version to 2.0.0-beta4

Update to beta3

Update environment.yaml

Merge pull request #255 from WrightonLabCSU/Viral_fix

Push a point release to react to ref-seqs changes

Fix config test

Merge pull request #228 from WrightonLabCSU/improved_amg_summary

Improved amg summary
This update changes the AMG summary so that it will provide all
information from both the Metabolic DB and The AMG DB. It is a bit of a
kitchen sink approach, but we can always cut it down if needed and most
dram-v users I would call advanced.
I could not fix the sql someone more familiar with sqlalchemy is free to do so
I will make a branch to deal with this

This is why we don't pass 30 vars by position

Make real headers

DRAM v1.3 change log

* Add --amg_database_loc parameter that was missing in DRAM-setup.py
* Shift DRAM download of UniRef from FTP to HTTP address to address firewall issues
* Rename of headers in annotations.tsv files to be more uniform across databases
* By default `DRAM.py annotate` now does not annotated with VOGDB by default, flag added to use VOGDB
* By default don't split `DRAM-v.py annotate` input contigs into separate files because HMMER doesn't care for E-values
* Users can now pass multiple `--input_fasta` arguments to `DRAM.py annotate` and `DRAM-v.py annotate`
* Now DRAM makes sure bin names (pulled from file names) are unique and not full paths
* Update pandas methods to get rid of warnings and increase speed
* When annotating with KEGG Genes the KEGG Genes IDs are stored in the `annotations.tsv` in addition to the KO IDs
* Complete rewrite of how HMM annotation is handled inside DRAM to reduce redundancy and allow...
* Users can now annotate using custom HMM sets which may include custom bitScore cutoffs
* Complete rewrite of database handling from setup through annotation, in the future this will allow more flexible configuration
* Change CI to CircleCI from travis
* DRAM strainer and gene neighborhood pulling can now both use custom distillate information