ENH: plot flight center-of-pressure evolution (#954) - #1130

we need to see one or two examples of the new plot before merging it please

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 82.83%. Comparing base (e0ff281) to head (8207282).
⚠️ Report is 75 commits behind head on develop.

@@             Coverage Diff             @@
##           develop    #1130      +/-   ##
===========================================
+ Coverage    82.18%   82.83%   +0.65%     
===========================================
  Files          122      129       +7     
  Lines        16355    16839     +484     
===========================================
+ Hits         13441    13949     +508     
+ Misses        2914     2890      -24     

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

@Gui-FernandesBR here are two examples from this branch:

1. Full-ish ascent (rail → burnout → apogee window), truncated at first_event_time like other stability plots:
   

2. terminate_on_apogee=True:
   

Both generated with flight.plots.center_of_pressure(filename=...) under MPLBACKEND=Agg. Happy to tweak axis limits / twin-axis layout if you want something closer to the existing stability plots.

the example plots don't really make sense

Following up on my "the example plots don't really make sense" comment — I found why, and it is not the plotting code.

static_margin is only ever discretized over the burn:

self.static_margin.set_discrete(lower=0, upper=self.motor.burn_out_time, samples=200)

With extrapolation="constant", every query past burn_out_time returns the value frozen at the edge of that range. So for the whole coast, apogee and descent the curve is not a computed static margin at all — it is one number repeated, which is exactly the flat/implausible shape in the plots you posted.

For a flight-long quantity, use Flight.stability_margin, which is a function of Mach and time and is evaluated over the whole flight. If you really want the static margin specifically, it has to be re-discretized over the flight's own time span before plotting.

Two other notes:

• The workflow runs on this PR had been stuck in action_required and never actually executed, so this PR has never had real CI. I approved them, so you will get results now.
• Once the curve is right, please post the regenerated plots again so we can compare.