Audit exact single-repair hardening effects

Extends the repair-loss frontier with an exhaustive counterfactual for every minimal repair: assume that one repair remains available, then enumerate every exact-size failure set through three unavailable repairs among all remaining repairs. Each candidate returns baseline versus hardened worst-case observation loss, avoided loss, exact remaining tie count, a deterministic witness, and repairs present in every or any remaining worst case. At most 70 candidates, 280 candidate/budget points, and 3,836,700 repair/failure-set comparisons are admitted. Results are exact for returned grid/sample evidence, not continuous-time guarantees, probabilities, causal infrastructure claims, rankings, recommendations, interpretations, predictions, or statistical confidence. No licensed catalog, new dependency, or request-time network source is used. One standard quota unit per admitted request; failed or timed-out admitted attempts are not refunded. Do not automatically retry. Historical mode uses an explicit BCE/CE date object and UT1 conventions. Its response is the versioned historical envelope, not the modern UTC envelope. Read result for calculation fields and metadata for limits and uncertainty.