Amptk

- ITS database has now gotten too large to be hosted in single file on OSF, so its split into two. All users will need to upgrade to v1.6.0 in order for database to be properly downloaded, used.

- bug fix for `amptk unoise3` 96 99

* bug fix for `amptk database` that was stripping species epithet names 81
* bug fix for linux systems related to platform, fix was to use `distro` 91
* added action to build docker images and push to docker hub. Fully functional version can be run with shell `amptk-docker` script, usage see https://amptk.readthedocs.io/en/latest/index.html#run-from-docker
* Updated fungal ITS database based off of UNITE v8.3, upgrade with `amptk install`
* Added new PR2 database for universal SSU amplicons, https://github.com/pr2database/pr2database, add with `amptk install`
* Re-wrote `amptk funguild` instead of running the original `Guilds.py` script as this stopped working at some point
*

* typo/bug fixes 84 85
* fix to multiprocessing so now py>3.7 working 83

Bug fix release

* fix for merging PE reads with `vsearch`
* fix dereplicate function in `database` 77
* fix for `filter` if `--negatives` passed but no mock community barcode
* fix for `taxonomy` where custom databases not working with the `-d` flag 80
* add `-p illumina3` to `SRA-submit`

* to support OSX Catalina dropping 32-bit applications, I've removed `usearch9` as a strict dependency. `vsearch` will be the default for all processing steps, including taxonomy assignment. Along these lines, I've dropped UTAX classifier as "default" in the hybrid taxonomy method.
* dropping usearch, required addition of a few new dependencies: `mafft`, `fasttree`, and the python package `pyfastx` (for speed simplicity).
* added support for PacBio CCS reads. Reads can be processed with `amptk pacbio` and then clustering with `amptk pb-dada2`
* several bug fixes
* fix embarrassing typos in v1.5.0 -- do not use v1.5.0 its broken.
* apparently it does not work in python 3.8 -- investigating that for future release/support.
* working on ONT (Oxford Nanopore) support as well -- stay tuned for this in near future.