Add read filtering and taxonomic classification documentation

[charlesreid1]

This adds the initial materials for an mkdocs documentation site.

The documentation index is currently juts a copy of the repository README.

We are creating a duplicate documentation structure (duplicating the workflow documentation already in workflows/), and revising these documents as they are migrated over into docs/. For example, we convert README to index. We copy workflow overviews into docs/ folder workflow documents. This creates a front-end (searchable documentation with mkdocs-material) that is the "finished product" documentation.

• [x] README.md (the first page the user sees when they visit the repo) and docs/index.md (the first page the user sees when they visit the documentation) will contain the same info
• [x] workflow documents in workflows/ --> docs/workflows/
• [x] revise/combine: workflow overview, then one page per workflow (descriptions of each workflow, and instructions on running workflows, location of files): start with read filtering and taxonomic classification
• [x] Keep it simple (don't "redo everything"), but bring the material up to date ("this tutorial covers docker commands, the Snakefile uses singularity directives...").
• [x] Snakemake rules notes: logic/difference between walkthroughs and Snakefiles
• [x] Keep the walkthroughs - valuable - but explain how differ from Snakefile commands ("many bothans died to bring us these rule files....")
• [x] calling a workflow (the Snakefiles): command line

Issues that we won't deal with:

• how to provide both Docker and Singularity Snakefiles? (leaving in the old docker Snakefiles as dockerSnakefile, but we need to figure out a better plan for these.)
• additional examples of how to call the Snakemake API from a script or a notebook? (leaving out, sticking to command line for simplicity)

Merge Checklist

• [x] Typos are a sign of poorly maintained code. Has this request been checked with a spell checker?
• (N/A) ~Tutorials should be universally reproducible. If this request modifies a tutorial, does it assume we're starting from a blank Ubuntu 16.04 (Xenial Xerus) image?~
• (N/A) ~Large diffs to binary or data files can artificially inflate the size of the repository. Are there large diffs to binary or data files, and are these changes necessary?~

[charlesreid1]

Remaining to do:

• Test this puppy out with latest singularity
  • [x] dahak-yeti install singularity script - more robust, take version number
  • [x] implement ghost build rules
• Fix header+comments in Snakefile
  • [x] use Phil's original author/header info
  • [x] remove try/except
• Documentation:
  • [x] create docs folder and dahak documentation (see #84)
  • [] ~workflow writeup for read filtering/taxonomic classification: these are the rules and brief descriptions~ (see #84)
  • [] ~where to put explanation of snakemake rules and gotchas?~ (see #84)
  • [] ~one custom snake file for user to use for shimming in their own rules~ (take care of this in the PR that implements Snakefile organization)

[charlesreid1]

Ready to merge.

Will also be able to use this to provide a final working documentation site at https://dahak-metagenomics.github.io/dahak once we merge this and build it.

[charlesreid1]

This pull request is ready, but it should happen together with #83, since it documents several changes made in #83.

[brooksph]

Review in progress

[brooksph]

Closed in favor of https://github.com/dahak-metagenomics/dahak/pull/95