We can run the testcase_full (a testcase that contains all the method calls of
the data structure) to infer all the parameters in one file, but that takes a
long time that is mainly spent on the model-checking for unknown wrong
inferences. To make things faster, we could split the big testcase into small
cases (plus corner cases) and then run each of them one after another.

We ran testcase1 and then got inference result1.txt, then we ran testcase2 based
on result1.txt, and we got result2.txt, which is the correct inference.

result1.txt -> 1m56.110s
result2.txt -> 0m0.207s

total -> 1m56.317s