Use random tensor sizes in tests

[wavepacket]

Just a comment to not forget it: There was recently a bug in trace() with incorrect calculation of indexes. This was not captured by the unit tests, because it only happened for unsymmetric tensor shapes, while the tests use only equal sizes for all dimensions.

So suggestion: When iterating over tensors of different ranks and sizes in the tests, use unsymmetric (random?) tensor shapes.

[juanjosegarciaripoll]

test_over_tensors() in loops.h now iterates over tensors of different sizes and ranks. Is this what you meant?

[wavepacket]

I am not sure. I can say for certain that the bug mentioned in my original posting referred to commit dfdef4d, but I cannot fathom what I originally meant here.