Add changes from Sarah's development branch

This PR includes development from the sarah-dev branch. This branch has been used for testing on Aspen and is working, both for SN and JG.

Changes:

[x] Allocate resources intelligently (previously each process got 1 core, regardless of process complexity or dataset size)
[x] Fix pigz dependency missing in container by using gzip instead; put file cleanup into its own process
[x] Fixed error with non-gzipped input sequences (symlinks weren't working with Nextflow)
[x] Added wrapper script and updated README for Aspen cluster users
[x] Improved version pinning and logging
[x] Check for trivial alignment length, a sign of too-unrelated input fed to ParSNP. Allow user to specify non-curated input.
[x] Store temp files (work directory) in /scicomp/scratch/$USER to avoid responsibility to purge temp files ourselves
[x] Stop copying input genomes to work/ directory, figure out how to make symlinks play nice with nextflow
[x] Match nf-core style (move bin/ scripts to script block, one module file per process, add test and test_full config profiles)
[x] Add github actions workflow to run full test for CI

Remaining TODOs:

[ ] Bring back pigz in modules/local/outfile_cleanup.nf (it's not in the ubuntu docker container, using gzip instead)
[ ] After PR is merged, swap out $LAB_HOME/workflows/wf-assembly-snps on Aspen for CG repo (currently it's SN's fork on sarah-dev branch)

chrisgulvik / wf-assembly-snps