Fix integration pipeline

This MR fixes two tests which fail in the pipeline. test_version_string because it was just broken and test_cumulant which failed due to timeout. This case was fixed by replacing a symbolic comparison between collision rules with a numeric comparison.

Furthermore, some long-running cases are parallelised now which should decrease the overall time a little bit and increases the readability when a test case fails since py-test will show the exact parameter setup now on failure.

lbmpy_tests/test_split_optimization.py

@@ -34,6 +34,7 @@ def test_split_number_of_operations():
 @pytest.mark.parametrize('compressible', [True, False])
 @pytest.mark.parametrize('method', ['srt', 'mrt'])
 @pytest.mark.parametrize('force', [(0, 0, 0), (1e-6, 1e-7, 2e-6)])
+@pytest.mark.longrun
 def test_equivalence(stencil, compressible, method, force):
     relaxation_rates = [1.8, 1.7, 1.0, 1.0, 1.0, 1.0]
     clear_cache()