Fix ggevidence plot bug

[pawelqs]

• fix: correct position calculation in ggEvidenceplot function
• tests: adds test of ggEvidenceplot function

[pawelqs]

Fixes #3

[january3]

Zrobione, wielkie dzięki!

Frankly I abandoned the ggEvidenceplot function because I was not able to implement gene labels due to the conceptual limits of the ggplot engine. You simply cannot know how label font sizes translate to user coordinates, so I would have to meddle with low level grid functions.

[pawelqs]

No problem ;)
Regarding the gene labels - I used to do similar things with ggrepel. Have you tried to solved that way?

[january3]

Nah, ggrepel doesn't work here, not if you have more than a couple of genes. Maybe I should put it back in, it is better than nothing (then again, the user can simply add the ggrepel() call to the object returned by ggEvidencePlot). New tmod release coming soon, but I need to fix one more bug first.

[pawelqs]

It does not scale well to hundreds of genes, but works pretty good for few dosens if the labels are aligned like this:
https://ggrepel.slowkow.com/articles/examples#align-labels-on-the-top-or-bottom-edge ;)