style(tracingfuncs): Numpydoc validation, PEP8 & tidying

[ns-rse]

Resolves #920

This commit is a squash of seven individual commits that address Numpydoc validation

• style(tracingfuncs): GL01/GL02 Numpydoc validation
• style(tracingfuncs): SS06 Numpydoc validation
• style(tracingfuncs): GL08 Numpydoc validation
• style(tracingfuncs): PR01/RT01 Numpydoc validation

Removes unused functions/methods...

• fix(tracingfuncs): Removes circularTrace_old
• fix(tracingfuncs): Remove unused getLocalPixelsBinary

Renames methods to adhere to PEP8

• style(tracingfuncs): PEP8 renaming camel-case method to snake-case

Questions

check_vectors_candidate_points - part 1

This function (nee checkVectorsCandidatePoints) takes the parameters x and y but these are never used directly, rather x and y are obtained anew in each iteration of this loop...

        for x_n, y_n in candidate_points:
            x = x_n - x_ref_2
            y = y_n - y_ref_2

            theta = math.atan2(x, y)

            x_y_theta.append([x_n, y_n, abs(theta - ref_theta)])

We should be able to just drop x and y as parameters to this method but I wanted to check as there aren't any tests for these methods/functions.

check_vectors_candidate_points - part 2

I also noticed that in some places check_vectors_candidate_points (nee checkVectorsCandidatePoints) was commented out (lines 83 and 187) in favour of using find_next_best_point (nee findNextBestPoint) but check_vectors_candidate_points is still used in some places.

• Is this deliberate?
• Can or should find_next_best_point replace all calls made to check_vectors_candidate_points?

[MaxGamill-Sheffield]

@ns-rse I have a rough idea what this piece of legacy code does but not how different things could be interchanged. I'd say leave this for now and get this adhering to the standards. And possibly make a new issue to swap these out and see if any tests are affected?

[ns-rse]

Looks good, just if you want to refactor the legacy code or make a new issue?

I've gone with somewhere in-between and addressed Part 1 as removing the unused parameters had no impact on the tests completing successfully and kicked the can done the road for Part 2 and written it up in #953