Context deadline exceeded error lacks accessible details in Prometheus UI

Goal: ensure that when a PromQL query exceeds the configured timeout, the error presented in the web UI conveys sufficient detail and is announced properly to assistive technologies, enabling users to understand the cause and adjust their workflow.

Currently the UI shows a generic 'Query failed' banner without ARIA labels or specific message, while the backend logs the 'context deadline exceeded' error. The timeout is set only via the global --query.timeout flag, with no per‑user or per‑request override, a design decision still under discussion. This leaves users on slow connections or relying on screen readers without clear guidance.

Should the UI display the underlying error string or a more descriptive message? Should ARIA live regions be added to announce the failure? Is there value in exposing a per‑request timeout mechanism via query parameters or HTTP headers?