fixing some pep8 details

[fchapoton]

Checklist

• [x] Added an entry in doc/news/. (Ok. None needed.)
• [x] Added a test for this change. (Not part of the CI yet.)

[fchapoton]

the failure happens on a line tagged "random", that should be tagged "not tested" and the next line too

[fchapoton]

ping ?

[saraedum]

@videlec we have to do something about these randomly failing Lyapunov exponents…namely:

 File "surface_dynamics/flat_surfaces/origamis/origami_dense.pyx", line 1826, in surface_dynamics.flat_surfaces.origamis.origami_dense.Origami_dense_pyx.?
Failed example:
    o.lyapunov_exponents_approx(involution=s, nb_iterations=2**21)  # abs tol .05
Expected:
    ([0.6000, 0.2000],
     [0.8000, 0.4000])
Got:
    ([0.450016509703521, 0.150057924057956],
     [0.599994117515633, 0.300090271196313])
Tolerance exceeded in 3 of 4:
    0.6000 vs 0.450016509703521, tolerance 2e-1 > 5e-2
    0.8000 vs 0.599994117515633, tolerance 3e-1 > 5e-2
    0.4000 vs 0.300090271196313, tolerance 1e-1 > 5e-2

[videlec]

@videlec we have to do something about these randomly failing Lyapunov exponents…namely:

 File "surface_dynamics/flat_surfaces/origamis/origami_dense.pyx", line 1826, in surface_dynamics.flat_surfaces.origamis.origami_dense.Origami_dense_pyx.?
Failed example:
    o.lyapunov_exponents_approx(involution=s, nb_iterations=2**21)  # abs tol .05
Expected:
    ([0.6000, 0.2000],
     [0.8000, 0.4000])
Got:
    ([0.450016509703521, 0.150057924057956],
     [0.599994117515633, 0.300090271196313])
Tolerance exceeded in 3 of 4:
    0.6000 vs 0.450016509703521, tolerance 2e-1 > 5e-2
    0.8000 vs 0.599994117515633, tolerance 3e-1 > 5e-2
    0.4000 vs 0.300090271196313, tolerance 1e-1 > 5e-2

I agree. One could ignore completely the values and just check consistency of the list sizes in the output?