As an example for how it is right now, have a look at the brunel-extraptester
test in :
The order of the algorithms and counters reported isn't fixed, which makes the comparison harder than it needs to be.
This MR tries to ensure we print these differences in alphabetical order.