Add small test dataset

[invertdna]

When you get a chance, can you add a small test dataset to the repo, so that new users can quickly establish if the pipeline (as a whole) is working? See, for example, Ramon's pipeline, which has a subset of a real dataset here https://github.com/ramongallego/Nextera_Dada2/tree/master/data_sub .

And then the example script can point to a known (small) dataset, do the analysis, and result in predictable pipeline output.