Improve device management and monitoring

[temiroff]

What changed

• Clarifies compute-device, robot, deployment, SSH, update, and lifecycle controls in the editor.
• Adds responsive Runtime + Robot Hardware reports and durable version reporting.
• Adds selectable live robot monitoring for idle Hardware and active deployments.
• Improves stored-deployment restart behavior, torque reporting, progress handling, and related documentation.

Why

Device ownership, service versions, deployment persistence, and physical torque were presented through overlapping controls and transient status. This made it difficult to tell which robot was targeted and whether a workflow remained installed.

Impact

Operators can manage each robot from a clearly separated card, inspect reliable service versions, restart stored deployments, and view live joint telemetry from the current hardware owner.

Validation

• python -m unittest discover -s tests (438 passed, 8 skipped)
• cd editor && npm run build
• Editor device API tests (70 passed)