This build starts from the integrated five-improvement viewer and applies the A/B-supported runtime-cost removals. The numerical architecture remains direct f32 / perturbation + series / sparse recovery / finite continuation / CPU fallback.
Historical investigation records remain under `docs/performance-cost-audit/` and `docs/integrated-improvements/`.
## Reported deep-view artifact fix (2026-09-29)
A reported deep URL exposed an unsafe 4th-order series jump and premature publication of FAST numerical failures. Production series acceleration is disabled pending a propagated error-bound redesign, and FAST provisional publication now occurs only after numerical failure repair. See `docs/reported-view-bugfix/IMPLEMENTATION.md`.