Add more details to annotation pipeline sample and browsing genomic data.

[deflaux]

Also:

• added skeleton content for CountReads Dataflow. We'll leave it unlinked from the TOC until @iliat has a chance to fill in.
• added some guidance about run times for dataflow and spark jobs.

@mbookman please take a look http://googlegenomics.readthedocs.org/en/staging/ thanks!