Enhance spoken alarm feedback and controls

[nedtwigg]

Summary

• remove asterisks from spoken terminal labels so they are not announced aloud
• track actual speech-engine delivery and retain a per-Session SPEAKING / SPOKEN state until the alarm is deliberately attended
• add loud, theme-adaptive Pane and Door treatments for active and completed speech
• expand the baseboard alarm controls into separate audio-status, push-status, and general settings buttons
• update the alert, layout, theme, and design documentation together with focused tests and Storybook scenarios

Why

Spoken alarms were difficult to associate with a terminal, and labels containing * produced confusing announcements such as “eight asterisk.” The settings affordance also hid whether speech and push delivery were enabled. This makes delivery state unmistakable while keeping resolution tied to deliberate attention rather than mere visibility.

Validation

• pnpm test
• pnpm --filter dormouse-lib build
• pnpm --filter dormouse-lib build-storybook
• node scripts/spec-lint.mjs
• Storybook visual QA in light and Monokai themes
• git diff --check

[cloudflare-workers-and-pages]

Deploying mouseterm with    Cloudflare Pages

Latest commit:	9333103
Status:	✅  Deploy successful!
Preview URL:	https://f26dd22f.mouseterm.pages.dev
Branch Preview URL:	https://speak-alarms.mouseterm.pages.dev

View logs