chore: prep for first PyPI release (v0.1.0)

[rramboer]

Summary

Prepares the project for its first publication on PyPI. Includes packaging-metadata fixes, tooling consolidation, several public-API improvements, and a number of correctness and documentation fixes.

Changes

Packaging & metadata

• Set version to 0.1.0; classifier Development Status :: 3 - Alpha
• Replace placeholder yourusername URLs with the real repository owner
• pyproject.toml is now the single source of truth for dependencies; remove requirements.txt and requirements-dev.txt

Tooling

• Consolidate linting and formatting on Ruff; remove Black, isort, and Bandit
• Add Python 3.13 to the CI test matrix (already declared in classifiers)
• Add pip caching to CI jobs and install dev extras (pip install -e ".[dev]") instead of hard-coding tool versions
• Pre-commit configuration updated to match new tooling

Public API

• simulate_rocket() and simulate_multiple() now accept an optional body: CelestialBody | None parameter, enabling Moon, Mars, or custom-body simulations from the convenience API
• Add a body field to Rocket; potential_energy, total_energy, and the gravity term in Rocket.update now use the correct body's surface gravity and radius instead of unconditionally using Earth
• get_preset() returns a dataclasses.replace() copy instead of the shared global RocketConfig instance, preventing accidental mutation of PRESETS
• Export CelestialBody and SimulationState in rocket_sim.__all__

Bug fixes

• Remove logging.basicConfig(...) call from RocketSimulation.__init__. Library code must not reconfigure the host application's root logger
• Remove unused Engine.specific_impulse field. It was stored but never read; the integrator uses a constant-mass approximation, which is now explicitly documented
• Use <= instead of exact float equality in Rocket.is_on_ground
• Correct erroneous example values in Physics.gravity_at_altitude and Physics.escape_velocity docstrings (verified against actual computed outputs)

Documentation

• Add explicit modelling caveats to the README: motion is 1-D vertical, mass is constant, and atmospheric drag is not applied to trajectories
• Document Physics.atmospheric_density and Physics.drag_force as standalone utilities, not part of the default trajectory loop
• Footnote the New Shepard preset to explain its sub-1 thrust-to-weight ratio at the listed masses
• Update CONTRIBUTING.md and CHANGELOG.md to match the new tooling and version

Removals

• Delete unused PlotConfig dataclass; it duplicated PlotOptions and was never imported. Removing it before the first publish avoids a breaking change later
• Remove dead IntegrationMethod enum and SimulationConfig.integration_method field, which were defined but never consulted by Rocket.update
• Remove SimulationConfig.log_level (only used by the now-removed basicConfig call)

Test plan

• pytest — 62 tests pass
• ruff check src tests — clean
• ruff format --check src tests — clean
• mypy src — no issues
• python -m build — builds rocket_sim-0.1.0.tar.gz and rocket_sim-0.1.0-py3-none-any.whl
• twine check dist/* — both pass
• Smoke test in a clean virtualenv: install the wheel, import the package, run a Falcon 9 simulation on Earth and on the Moon, invoke the rocket-sim CLI
• Verify get_preset() does not leak mutation into the global PRESETS registry
• CI green on this PR