Corrected overflow with ele_stop <= ele_start

[freddieknets]

Description

Using ele_stop in track() implies that the last turn will be incomplete (unless ele_stop=0); hence no extra turns should be added.

I.e. if we do track(num_turns=2, ele_start=10, ele_stop=20) the first turn tracks from element 10 to the end, and the second turn from the start until element 20 (and similarly for e.g. num_turns=3 where the middle turn is a complete one).

In this case, nothing changes when ele_stop <= ele_start:
If we do track(num_turns=2, ele_start=20, ele_stop=10) the first turn tracks from element 20 to the end, and the second turn from the start until element 10.

There is one important exception: when ele_stop <= ele_start and num_turns=1. This is in principle impossible, but writing track(ele_start=20, ele_stop=10) should be a convenient way to track only a part of the line (that passes through the origin) without having to worry about increasing num_turns manually. Hence, for this very specific overflow case, and only for this one, the tracker changes num_turns to 2.

This was the original idea, however, the overflow was applied to all values of num_turns which creates a discrepancy in physical interpretation between the cases ele_stop <= ele_start and ele_start < ele_stop.

Checklist

Mandatory:

• I have added tests to cover my changes
• All the tests are passing, including my new ones
• I described my changes in this PR description

Optional:

• The code I wrote follows good style practices (see PEP 8 and PEP 20).
• I have updated the docs in relation to my changes, if applicable
• I have tested also GPU contexts