Disallow nondistinct variable names for universal rings

[SoongNoonien]

This is the first step in the direction of #2408.

[SoongNoonien]

I'm not sure if this should be tagged "breaking". While it technically is breaking, it shouldn't have been possible in the first place to create a universal ring with duplicate variable names.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.12%. Comparing base (2657ac7) to head (5875ae8).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2413      +/-   ##
==========================================
- Coverage   88.13%   88.12%   -0.01%     
==========================================
  Files         130      130              
  Lines       32948    32949       +1     
==========================================
- Hits        29038    29037       -1     
- Misses       3910     3912       +2     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[fingolfin]

@fieker is requesting a new interface akin to add_vars(UnivR, num) which adds num new variables to a universal ring, and returns them. I.e., it takes care of choosing admissible names.

I think this can be done in a follow-up PR.

[fingolfin]

Regarding the breaking label: I'd be fine with just merging this, as long as it doesn't break our upstream tests. If it breaks something for another user, I am sorry, but most likely there already were issues in their code. If someone is affected by this and reads this: sorry! but please get in touch with us, then we can make sure next time we are aware of your needs and take them into account.