Feature/belated pep8 compliance

[jackraymond]

Pull requests 51, 45 (to predecessor dwave-neal), and 25 introduced a large number of PEP8 violations, this pull request corrects those errors (with some small amount of compliance collateral not related to the pull requests).
In this pull request I made several files code complaint with pep8 and pyflakes3.

I plead guilty to charges E501, E265, E231, E261, E262, E251, E225, E112, W293 and others. Now that I program primarily in python I see the ugliness.

I standardized sa/sampler.py and tabu/sampler.py at width <=79 wrap around (with a few very difficult to compress exceptions at <99). The code was close to <=79 convention in most places, so this seemed sensible.
For test_simulated_annealing_sampler() and test_tabu_sampler() I allowed many lines <=99 to remain as written, except where local style consistency dictated their shortening.

The changes for the most part errors I introduced in pull requests, in a handful of cases they potentially override stylistic choices of other contributers. Apologies for this! Also apologies to my main reviewers the last 2 years.

[codecov]

Codecov Report

Merging #56 (51bc6ae) into main (f412d64) will increase coverage by 1.13%.
The diff coverage is 80.95%.

@@            Coverage Diff             @@
##             main      #56      +/-   ##
==========================================
+ Coverage   93.02%   94.16%   +1.13%     
==========================================
  Files          17       17              
  Lines         545      531      -14     
==========================================
- Hits          507      500       -7     
+ Misses         38       31       -7     

Impacted Files	Coverage Δ
dwave/samplers/sa/sampler.py	88.88% <73.33%> (+3.92%)	⬆️
dwave/samplers/tabu/sampler.py	98.07% <100.00%> (+0.07%)	⬆️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[arcondello]

For test_simulated_annealing_sampler() and test_tabu_sampler() I allowed many lines <=99 to remain as written, except where local style consistency dictated their shortening.

It looks like a lot of lines with <100 characters got changed to <80. IMO it would be better to stick to <100 and minimize the changes. Also, 100 is (IMO) much more readable on modern systems than 80.