mathInfo: unset undefined (None) attributes

[anthrotype]

see conversation at #176 (comment)

[codecov]

Codecov Report

Merging #177 into master will increase coverage by 0.06%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master    #177      +/-   ##
=========================================
+ Coverage   88.13%   88.2%   +0.06%     
=========================================
  Files          12      12              
  Lines        2284    2297      +13     
  Branches      303     302       -1     
=========================================
+ Hits         2013    2026      +13     
  Misses        193     193              
  Partials       78      78

Impacted Files	Coverage Δ
Lib/fontMath/mathInfo.py	85.3% <100%> (-0.07%)	⬇️
Lib/fontMath/test/test_mathInfo.py	100% <100%> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 519da52...d59c1fa. Read the comment docs.

[justvanrossum]

I definitely agree with the test changes, and I think I also agree with the setattr change. Let me think out loud: the only thing this really changes is that when number lists have unequal lengths the resulting field will be set to None instead of taking the value from the first operand. Since that can't really have had meaningful results, I think this means this change is ok. Is my reasoning correct?

[anthrotype]

the only thing this really changes is that when number lists have unequal lengths the resulting field will be set to None instead of taking the value from the first operand.

correct. I added a test for that too.