Kneaddata

* Fixed trimmomatic conda executable
* Added the hg39 T2D Human genome reference database
* Fixed TRF parallel run bug
* TRF broken download links updated

* Added dog reference genome database
* PyPi + conda release updated

* Allow for tracking of paired-end reads from bam input files
* Add an option to use a scratch directory

* Change the input format to --input1 --input2 --unpaired
* Trim the spaces in the identifiers sequences of the input files
* Add the identifier sequence number if not present in the input files
* Allow for files with less then N lines (in checks for sequence identifiers)

* Remove the file sort from the function that counts the reads so the file name and database name are always in sync. Resolves issue with workflows where read counts are out of sync for database names.
* Change trimmomatic min read length to 60 nt.
* Add back in legacy option "--run-trf" for backwards compatibility with workflows.