Provide an option to incrementally write Gene Finder output to disk

[alexismhill3]

Also adds another closely related feature: the pipeline can now process data starting from an arbitrary contig in a file.

This makes a couple changes to the Gene Finder API that are not backwards compatible. Basically, I really didn't like how some arguments were basically redundant, but I couldn't think of a way to fix this without leaving a bunch of clunky, non-functional arguments. Since we currently only have 3 users I figure it's not a huge problem. Anyway, here are the changes:

• outfrmt is no longer a parameter of pipeline.run; only the CSV format that we have been using will be supported moving forward.
• outfile and all_hits_outfile are also deprecated. Instead, the user can specify a directory to write all output files to with output_directory, and job_id can be used to prefix output files.

I manually tested this on some real metagenomic data and it seems to be working.

[alexismhill3]

Okay this should be ready to be reviewed again