testdata/expected/results-offline.csv is quite big

[tales-aparecida]

testdata/expected/results-offline.csv was already kind of big and you've 4x it, into 12k lines (1.3MB). Any way to break it into smaller files? If not in this MR, at least a plan on how we could achieve that

I have considered that.
The main problem here is, in the current implementation the file diff between expected and result is used as a strategy to analyze differences.
Breaking into per test files might help reducing the big file problem, but I am not sure this is a good solution here.
Do you have any suggestions?

Originally posted by @alanpeixinho in #33 (comment)

[tales-aparecida]

@alanpeixinho created this to think about it later