fix the sign of Flight.potential_energy

[Mohit-Ak]

Pull request type

• Code changes (bugfix, features)

Checklist

• Tests for the changes have been added (if needed)
• Lint (ruff check / ruff format --check) has passed locally
• All tests (pytest tests/unit tests/integration) have passed locally

Current behavior

Flight.potential_energy computes mu * m * (1 / (z + R) - 1 / R), which is negative for z > 0 and gets more negative as the rocket climbs. Flight.total_energy inherits the sign error. Between burn out and apogee the mass is constant, so the change in potential energy should be m * g * dz; with the issue's script it comes out as -398754.6 J instead of +398711.0 J. The trajectory itself is not affected.

New behavior

The two terms are swapped, so the potential energy relative to sea level is positive and increases with altitude. Added test_potential_energy_increases_with_altitude, which checks that the burn-out-to-apogee change in potential_energy matches m * g * dz on the flight_calisto fixture. It fails on develop (-398045 J vs +398623 J) and passes with this change.

Breaking change

• No

potential_energy and total_energy change sign (they were wrong before); the energy plots in Flight.plots and CompareFlights will now show the expected curves.

Fixes #1202