Task 197: Fix 3D Orbital Analysis Dates and Data Provenance

Status: DONE

Claimed by: Claude Opus session

Problem

Deferred issues from Task 196 systematic review:

1. 3D orbital analysis dates wrong: Earth arrival in 3d_orbital_analysis.json is 2242-02-10, but the story timeline ends 2242-12-28 (479 days). The 3D analysis uses only 158 days for the full journey — ~10.5 months too short.
2. Plane-change fraction denominators wrong: orbital-3d-analysis.ts uses 2 * transferVelocityKmS as denominator for plane-change percentages, but these are approximate cruise velocities, not actual brachistochrone ΔV. Result: percentages are ~2x what they should be relative to total ΔV budget.
3. EP03 moon comparison ΔV values have no provenance: Values like Titania capture ΔV 0.37 km/s and orbit insertion 1.88 km/s appear in ep03.md but not in any calculation file. Need reproducible calculation script.

Plan

1. Fix leg dates in orbital-3d-analysis.ts to match actual episode timeline
2. Fix plane-change fraction denominator to use actual brachistochrone ΔV
3. Add EP03 Uranian moon ΔV calculations to recalculate script
4. Regenerate 3d_orbital_analysis.json and update cross-episode.md
5. Run all tests, verify CI

References

• reports/data/calculations/3d_orbital_analysis.json
• ts/src/orbital-3d-analysis.ts
• reports/data/summary/cross-episode.md
• reports/data/episodes/ep03.md
• reports/data/calculations/ep03_calculations.json