Reviewer badge shows incorrect number

re: - #1389 - #1397 It looks like @arfon did a fix for this once in #1397 , but it looks like review counts are still incorrect, e.g. here: https://joss.theoj.org/papers/reviewed_by/@majensen it should be 4 but currently the badge displays 1. Looking at each of the four papers, it looks like `majensen` is the first reviewer in one of them, but not first in all the others, could that be the bug? writing a test to check this out now.