feat(desktop,doctor): live Status dashboard + receiver controls

[pyramation]

Summary

wavegrid doctor becomes a live screen in the app instead of a terminal command — and the diagnostic logic moves out of the CLI so both render the same snapshot.

New @wavegrid/doctor (moved out of packages/cli/src/commands/doctor-checks.ts, which no longer exists): the pure checks, the two probes, and one collector that returns diagnostics as data:

const diag = await collectDiagnostics({ store, project, resolved, serverUrl? });
// { checks, overall, devices, sync{enabled,revision,entryCount,divergent,fromServer},
//   server: SystemStatus | null, serverUrl, serverError }

serverError distinguishes not-running (nothing on the port) from unauthorized / timeout / refused / not-wavegrid, which is what makes "brain down" actionable. runDoctor is now only rendering — plus the one case the collector can't express (no project to diagnose). --json now emits the collector's shape.

Desktop Status route. doctor:report adapts the snapshot in main (range/SystemStatus formatting stays out of the renderer, so no node deps cross the bridge). Layout follows the lesson from the Lights debugger: fixed-height page, two independently scrolling columns, nothing below the fold on a short window. Left = the show (brain version/layout/mode/uptime, connected receivers with shard + layout-mismatch flag, coverage gaps/overlaps, registered devices, sync divergence); right = this laptop's checks, sorted failures→warnings→passes, each with its remedy. Refreshes every 5s with a Live/Paused toggle.

Receiver controls. The receiver reads its OSC target, shard, and light map at startup, so a config change previously needed a full brain restart. brain.ts now owns the receiver handle independently:

startLocalReceiver()  // dials into the already-running brain
stopLocalReceiver()   // server + laser UI + connected clients stay up

Restart in the UI is stop+start. BrainStatus gains receiverRunning.

No protocol change: the report is read with the existing system_status request over an authenticated socket, exactly as the CLI already did.

Link to Devin session: https://app.devin.ai/sessions/972698f89f494b86828010666a002b8f
Requested by: @pyramation

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring