Outrageously long test in cyclic covers

[orlitzky]

Edit: deleted the first one which isn't reproducible in CPU time
Edit: and also the second one, which is #35479

File "src/sage/schemes/cyclic_covers/cycliccover_finite_field.py", line 1140, in sage.schemes.cyclic_covers.cycliccover_finite_field.CyclicCover_finite_field.frobenius_polynomial
Warning, slow doctest:
    CyclicCover(3, h).frobenius_polynomial()  # long time
Test ran for 1046.50 s, check ran for 0.00 s

And with CPU time (on a different machine):

File "src/sage/schemes/cyclic_covers/cycliccover_finite_field.py", line 1140, in sage.schemes.cyclic_covers.cycliccover_finite_field.CyclicCover_finite_field.frobenius_polynomial
Warning, slow doctest:
    CyclicCover(3, h).frobenius_polynomial()  # long time
Test ran for 248.93s cpu, 250.58s wall
Check ran for 0.00s cpu, 0.00s wall

[mantepse]

Could it be that there is a performance regression?

[orlitzky]

The main problem is that we aren't measuring how long the tests take in a meaningful way, but that will hopefully be fixed soon in #36226

I don't think there was a regression, I think they were always this slow.

[DaveWitteMorris]

Related issue: I think #35479 is the silver medalist.

[orlitzky]

Related issue: I think #35479 is the silver medalist.

Thanks, I updated the title/description. (Now that cputime is reported by default, anyone can look at the CI annotations to get a list of slow tests. I think I was extra frustrated when I reported these.)