Funannotate

- re-organize output so that temporary folders are created in the "final" resting place and not in the current directory
- modification of the multiprocessing function to include a simple progress percentage output
- bug fixes in `funannotate compare` and the orthology dN/dS output hanging when the dN/dS calculation failed
- modification to the logging to capture STDERR/STDOUT from many external tools into the log file, hopefully this will result in catching more errors than piping them to os.devnull
- bug fix in `funannotate annotate` where output folders not being created if the input was GFF, proteins, and fasta.
- made it a requirement to pass --species argument to `funannotate annotate` if you do not pass in a GenBank file. This is to prevent downstream problems in `funannotate compare` with how the scripts name the genome/isolates.
- made a [FAQ](docs/faqs.md) section in the docs that includes how to manually adjust gene models using the included tools
- all internal tests passed, however guaranteed there are more bugs. please let me know when you find them.

- update to dN/dS function to have two options: 1) --run_dnds estimate (which runs the M0 model only), and 2) --run_dnds full (which runs M0, M1, M2, M7, M8 and calculates the LTR of M1/M2 and M7/M8).
- update to the multiprocessing progress function - a simple progress meter is used on most multiprocessing functions to let user know how many processes have finished
- bug fix where transcripts and proteins were getting written to same file in `funannotate predict`
- minor bug fix in `funannotate clean` where input number of scaffolds was not printed out correctly
- change the default location of DB as per requested by some users, now defaults to $HOME/funannotate. Note you can set this to whatever directory you want.

- bug fixes for `funannotate compare`
- bug fix for `funannotate predict` during gene model filtering of large genomes occasional parent:child features would get missed
- added feature of calculating dN/dS ratios in `funannotate compare`

- integration of BUSCO2 script and models. Can see the BUSCO2 distribution [here](https://gitlab.com/ezlab/busco), funannotate uses a slightly modified version to be compatible with the BUSCO->EVM workflow.
- BUSCO2 models have changed a bit, there are now a lot more options for various taxonomic groups. Something to keep in mind though is that the model names for `dikarya` are not the same as `pezizomycotina` so if you use an `--outgroup` option be sure that the outgroup was generated with same BUSCO DB
- The `funannotate setup` script will remove previous BUSCO DB models and download the new ones because of the extensive change in the BUSCO2 structure.
- addition of `funannotate outgroups` to help you mange the outgroups available to `funannotate compare`
- the scripts will download and format any of the available BUSCO2 eukaryote models, to see a list in a taxonomic tree format you can type `funannotate outgroups --show_buscos`

- `funannotate annotate` will now support a single XML file for InterProScan5, you either pass a folder of single XML files 1 per protein, or a single XML file containing all of the annotations to the `--iprscan` option
- fix in `funannotate annotate` that did alert user that `--email` is required if using remote IPR5 search; this is default setting

- bug fix for path issue when running EVM; discovered on new install on Mac - not sure why it wasn't found earlier, but resulted in failed EVM run
- some improved logging for EVM module