add number_of_generators for UniversalPolyRing

[SoongNoonien]

It seems like the poor universal polynomials were overlooked again ;-) I tried to make this somehow in line with the MPoly implementation. In the documentation examples for the MPolys we are testing number_of_generators for equality instead of simply printing the result like in the new examples for UniversalPolyRing. Is there a particular reason for this?

[lgoettgens]

Thanks for your PR. At first glance, this looks good. But since we don't have the compatibility tests with Hecke and Oscar running at the moment (AbstractAlgebra already jumped to the next minor version, but downstream hasn't been adapted), I would like to wait (a few days) until this is working again, just to make sure to not accidentally break something.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 88.15%. Comparing base (ed58b14) to head (44a38d1).
Report is 11 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1795      +/-   ##
==========================================
+ Coverage   88.01%   88.15%   +0.14%     
==========================================
  Files         118      119       +1     
  Lines       30043    31271    +1228     
==========================================
+ Hits        26442    27567    +1125     
- Misses       3601     3704     +103     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[fingolfin]

I find it a bit weird to ask for the number of generators of an universal polynomial ring given that it is can change at any time. But if one takes it as an enhanced version of ngens(R) = length(gens(R)) then the code in here of course is fine. Also we already do this for number_of_variables, so...