docs: vision for extraction coverage — message-granular ledger, token-batched runs (2026-09-15 ideation)

The extraction pipeline stops re-reading whole sessions: coverage is a
per-chat covered-fingerprint ledger at MESSAGE granularity (tool call and
result never separated — they share a message; quiz events not special —
they ride re-opened messages normally), uncovered content is derived at
read time (never dual-tracked), the trigger becomes uncovered tokens >= a
chunk budget (fixing quiz-less sessions that never extract), and runs
checkpoint coverage incrementally per merged batch — abort/error loses
at most one in-flight batch. Manual extract-the-rest falls out of the
diff. Pointer line under vision.md 'Construction'.

docs/vision.md

@@ -39,6 +39,8 @@ Every node records every **(chat, message) point** where it came up; every edge
 
 ### Construction: free-form extraction, fully decoupled from sessions (DECIDED 2026-09-02 — supersedes live emission)
 
+*Coverage settled 2026-09-15 — see `vision-extraction-coverage.md`: the message is the unit of coverage (a covered-fingerprint ledger per chat), runs are token-budgeted batches of the oldest uncovered messages, and the trigger is uncovered tokens ≥ a chunk budget (replacing everyNQuizAnswers — quiz-less sessions extract too).*
+
 **Second amendment, reversing the 2026-08-22 live-emission decision.** The session's plan tree is the teacher's scaffolding for the session experience (§2); the global graph does not depend on it — structurally or otherwise. What the teacher formalizes is session-relative; the graph must not inherit those choices. (The original design doc's post-hoc stance was right after all.)
 
 Three stages, one pipeline — the sole graph writer: