Measure World forecast calibration

[eamirian]

What changed

• adds an exact World-owned public-event forecast record over the real FCC correction pair from P2C7
• issues treatment probability 0.75 and control probability 0.25 from the same original Observation before any correction identity or result material is available
• completes two reviewed treatment Actions and two reviewed control Actions before the held-out correction Observation is admitted
• exact-loads the later correction, derives the binary event from its explicit link to the original document, and derives single-event Brier quality rather than accepting a caller score
• records treatment quality 0.9375 and control quality 0.4375 over two matched pairs
• classifies the bounded difference as useful and emits only a non-effective, non-selectable promote proposal
• adds leakage, future-basis, injected-result, probability-bound, invented-score, replay, and fresh-host controls
• records a bounded work packet, immutable point-in-time audit, and restrained candidate roadmap state

Why

ACE 0.6 needs exact forecast/result provenance and calibration-sensitive outcome evidence in addition to citation correctness, contradiction attention, correction delay, and revision stability. This packet proves the unchanged Core + Intelligence measured-impact contracts can score a probability only after an exact held-out result becomes available.

This is stacked on World PR #11 and the Core measured-intelligence stack (#88, #89, #90). It does not merge, release, apply the proposal, pass SI4, close Core issue #38, or claim ACE 0.6 complete.

Verification

• focused P2C3-P2C9: 25 passed
• full World suite against the stacked Core candidate: 108 passed
• connector suite: 26 passed
• release contract: 7 passed
• public ace-core==0.5.0 boundary: 83 passed, 25 explicit candidate skips
• scoped Ruff check/format and git diff check: pass
• package build: unchanged 0.9.0 sdist and 45-entry inert data-only wheel; no Python or entry points; ace-core>=0.5.0,<0.6 unchanged

Repository-wide Ruff remains an inherited blocker. Using the same locked Ruff version, both the P2C8 parent and this branch report 14 lint findings and 12 format targets.

Claim boundary

This is a single-event Brier contribution over declared fixture probabilities and a held-out recorded result. It is not a historically contemporaneous forecast, an empirical calibration curve, a population reliability estimate, probability generation by ACE, a model-skill finding, live network monitoring, source independence, causality, general Brief quality, legal truth, or human benefit. The proposal is not applied.

Core issue #49 F1, F3, and F5 still need explicit 0.6 release-owner disposition; this packet does not expand scope to implement or re-date them.

[eamirian]

Superseded by the direct-to-main convergence in #17 and the published World Intelligence 0.10.0 release. The release lineage composes the P2C3–P2C10 work and preserves its evidence; this stacked draft is now closed for repository hygiene.