agent: lumotia — v0.1 release-completion run

Closes the code-side v0.1 ship gate. All quality gates green:
cargo fmt/clippy/test (~327 tests), npm check (0/0), vitest 13/13,
scripts/dogfood-rebrand-drill.sh 8/8.

Phase F — first-run onboarding promoted to v0.1
- FirstRunPage with skip-to-main + failure recovery + event recording
- Six onboarding commands (record/list/has-completed + lumotia_events)
- Storage migration v17 (onboarding_events + lumotia_events tables)

UI hardening (in-scope items from v0.1-ui-hardening.md)
- StatusPill + PostCaptureCard components, 21st preview entry
- Sidebar recording-as-sacred-state (opacity + aria-disabled, reduced-motion)
- Settings 6-section regroup + Help section + Activation log + Privacy toggle
- Error-state copy sweep (DictationPage + SettingsPage, plain-language)
- Global :focus-visible rule, textarea outlines restored
- Ctrl+K / Ctrl+, / Escape bindings in +layout

LLM resilience
- rule_based_extract_tasks (regex-free imperative-verb extractor) +
  extract_tasks_with_fallback wrapper — task extraction never returns zero
- tokio::time::timeout(120s) wraps cleanup/tags/tasks commands

Release artefacts
- LICENSE (canonical AGPL-3.0), CHANGELOG (Keep-a-Changelog format)
- v0.1-release-notes, privacy-and-ai-use, install-warnings,
  tester-onboarding-kit, tester-acceptance-runbook, code-signing-setup,
  apple-silicon-rb08-runbook, virtual-audio-setup, v0.1-contrast-audit
- Workspace versioning + AGPL spdx; npm exact-pin (10 ranges removed)
- AppImage SHA-256 sidecar in build.yml
- README v0.1 section + Reporting-issues; canonical repo slug

Closure pass — items moved from human-required to code-complete
- KI-02 Linux idle inhibit: zbus 5 → org.freedesktop.login1.Manager.Inhibit
- KI-03 Windows sleep prevention: SetThreadExecutionState(ES_CONTINUOUS|...)
- acquire/release_idle_inhibit Tauri commands, wired in DictationPage
- Diagnostic-bundle frontend wire-up (Settings → Help button)
- WCAG-AA contrast fix via .btn-filled-text utility (no token changes)
- 8 destructive-action sites wrapped in plain-language confirm() guards
- KNOWN-ISSUES.md + v0.1-known-limitations.md updated (KI-02/03 fixed)

Scripts
- pre-tag-verify.sh, tag-day.sh, smoke-linux + driver
- parse-diagnostic-bundle.sh, parse-activation-log.py

Per-item audit trail: docs/release/v0.1-completion-status.md
Remaining: W-01…W-08 (signing certs, hardware probes, smoke matrix,
tester recruitment) — see docs/release/v0.1-known-limitations.md.

docs/release/tester-acceptance-runbook.md

@@ -0,0 +1,39 @@
+---
+name: tester-acceptance-runbook
+type: release
+tags: [release, v0.1, tester-acceptance, runbook, 10-step, warm-activation, cold-setup]
+description: "Per-step runbook for the Lumotia v0.1 10-step tester acceptance flow. Documents expected outcomes, failure modes, and time budgets. Audience: developer walking through the flow personally on Linux, or an external tester on any supported platform."
+---
+
+# Tester acceptance runbook
+
+Expands the 10-step list from `docs/release/v0.1-checklist.md` into a check-off format with expected outcomes. Use this when walking through the flow personally on Linux, or to hand to an external tester.
+
+**Cold setup pass (steps 1–4):** no hard time bound — model download is the dominant variable.
+
+**Warm activation pass (steps 5–10):** target is 3 minutes once the model is ready.
+
+---
+
+| # | What to do | Expected outcome | Failure modes | Time |
+|---|---|---|---|---|
+| 1 | Download the artefact for your platform and install it. Linux: make AppImage executable, run it. macOS: open `.dmg`, drag to Applications, "Open Anyway" if Gatekeeper prompts. Windows: run `.msi`, click through SmartScreen. | App launches. No unusual escalation beyond the OS install prompt. | Gatekeeper/SmartScreen block with no bypass option. Installer fails. Linux AppImage reports missing FUSE. See `install-warnings.md`. | 2–5 min |
+| 2 | Launch Lumotia. | First-run onboarding screen appears; main UI not yet visible. | App skips onboarding and opens main UI. Blank window. Crash. | < 30 s |
+| 3 | Grant microphone access via the onboarding prompt. | Onboarding advances. No error about microphone access. | OS dialog does not appear. Granted but onboarding stalls. App reports denied after approval. | < 1 min |
+| 4 | Accept the suggested default model or choose another. Wait for download if needed. | Model selected. Download progress shown, completes cleanly. Onboarding advances to test-recording step. | Download stalls or errors. No default highlighted. Model picker missing. | 1–10 min |
+| 5 | Press Record. Speak for 10–30 seconds. Press Stop. | Recording starts immediately. Audio-level indicator is visible. Stops on command. | Record button unresponsive. No audio-level indicator. App hangs after stop. | < 2 min (start of warm-activation window) |
+| 6 | Observe the transcript pane after stop. | Raw Whisper transcript appears, matches what was spoken. | Pane empty. Spinner runs indefinitely. Error in place of text. | 5–30 s |
+| 7 | Observe the cleaned transcript tab alongside the raw one. | Cleaned version appears. Raw transcript still accessible. Cleanup failure shows a plain-language message; raw text preserved. | Cleaned tab blank. Raw transcript disappears. Stack trace shown. | 5–60 s |
+| 8 | Press "Extract tasks" in the post-capture card. If transcript had no task-like content, try again with "I need to email Alex by Friday". | At least one extracted task appears in readable text. | No tasks extracted despite task content. Spinner runs indefinitely. App navigates away before result visible. | 5–30 s |
+| 9 | Select an extracted task. Open MicroSteps. Add one step. Start the 5-minute timer. Navigate away and back. | Timer counts down visibly. State survives navigation. | MicroSteps panel does not open. Timer resets immediately. State lost on navigation. | < 2 min |
+| 10 | Go to History. Search for a word from the transcript. | Recording appears in results. Clicking it opens the full transcript. | Search returns nothing for a word that is in the transcript. History page empty. Crash on navigation. | < 1 min |
+
+---
+
+## Summary
+
+- [ ] Cold-setup pass (steps 1–4) — completed without coaching: yes / no
+- [ ] Warm-activation pass (steps 5–10) — completed in 3 minutes or less: yes / no
+- [ ] Tester confidence — would they use this again next week: yes / no / unsure
+
+Cold-setup: confusion counts as a failure, not just crashes. Warm-activation: model-inference latency is not a UX failure; unclear buttons are.