Run two gastric samples for bulk RNA-seq

Related to a bullet point from [this comment](https://github.com/AlexsLemonade/training-modules/issues/240#issue-647525460) on the potential bulk RNA-seq changes inspired by 2020 May and 2020 June trainings issue #240,

We should run two gastric samples for the bulk RNA-seq material since everyone is now similarly equipped using RStudio Server, and participants would typically have more than one sample that they want to process, thus giving them a better starting point.

Also from that bullet point, we should consider
> using a Makefile instead of a bash script for this per advice someone gave @cgreene

**Edit:** We should consider using Snakemake instead of a bash script.

Start with the linked comment and issue #240 to understand the proposed bulk RNA-seq changes. The issue mentions running two gastric samples and replacing the bash script with either a Makefile or Snakemake, but names no files or tests. Done means the training material supports the agreed sample count and workflow approach.

Written by the indexing model from the issue text.