docs: three pages cite closed issues for work that is unfinished or retracked #524

Area: docs — accuracy · found via docs review on #522

Three published pages point readers at closed issues for work that is either unfinished, retracked elsewhere, or explicitly not planned. Same defect class as #509 and #523: prose that outlived its subject and now hands the reader a dead tracker.

All three are in files #522 doesn't touch, so they're filed rather than folded in — #522 is already carrying two out-of-scope corrections.

1. docs/src/content/docs/durability.md:90 — promises a CLI that isn't tracked

"A self-contained wavehouse storage-check preflight subcommand that bakes this measurement and verdict into the binary … is tracked in #84."

• docs(deploy): JetStream SyncAlways=true durability vs throughput on commodity storage #84 is CLOSED/COMPLETED, and was never a feature issue — it's titled "docs(deploy): JetStream SyncAlways=true durability vs throughput on commodity storage". Its own comment says "mq: expose JetStream sync_interval as a config knob (throughput vs durability tradeoff) #139 = code-side knob; docs(deploy): JetStream SyncAlways=true durability vs throughput on commodity storage #84 = 'when and why' docs". It was closed by writing that page.
• cmd/wavehouse/ contains only main.go, health.go, validate.go — there is no storage-check, and no open issue tracks one.

Fix: either file the feature issue and re-point, or reword to say plainly that it isn't implemented and isn't currently tracked. Right now the page promises a preflight command that nobody is working on.

2. docs/src/content/docs/ingest-pipeline.md:258 — a whole section tracked to a shipped refactor

The ## Deferred / not yet implemented section is headed "Tracked under #191:", but #191 is CLOSED/COMPLETED — it's the refactor that shipped. The four bullets are still substantively accurate; only the tracking is wrong, in three different ways:

Also fix the code side: internal/ingest/worker.go:223 carries TODO(#191): tableLoops are spawned per distinct table and never … — the same dead pointer. Fixing the doc alone leaves half the drift.

Fix: re-point per bullet (#263, #264), and file or drop the two untracked ones.

3. docs/src/content/docs/sdk/streaming.md:115 and :188 — cites an issue closed as not planned

Both cite #449 for the live-query backfill-seam dedup defect. #449 was closed NOT_PLANNED on 2026-08-25 with: "Duplicate of #396. Same bug, same lines … Everything unique to this issue has been ported to #396."

#396 (bug(sdk): liveQuery dedup boundary is wrong — duplicate and dropped events in the core streaming feature) is open and is the superset.

The surrounding prose is accurate — only the pointer is dead. But "not planned" is the worst possible signal to send a reader about a live bug in a core streaming feature.

Fix: swap both references to #396.

Related: #523 (same review pass, OTLP port claim), #521 / #522 (the WH001 work that surfaced these)