fix(app): time out requests the server never answers

[Hona]

A socket that dies while a phone sleeps can leave fetch waiting for response headers until the OS gives up on TCP retransmits, which is minutes on Android and Linux. The request queue allows four in-flight requests per server, so a few of those dead requests wedge every later API call, and the only symptom is a server thrashing detected console warning.

• Each queued request now gets a 60 s headers deadline. If no response arrives by then the request is aborted with a TimeoutError and its slot is released.
• The deadline covers headers only; once the response arrives the body may stream for as long as it needs.
• Caller AbortSignals still propagate to the underlying request.
• /api/event keeps its existing bypass; the event stream has its own idle watchdog.

const controller = new AbortController()
request.signal.addEventListener("abort", () => controller.abort(request.signal.reason), { once: true })
const timer = setTimeout(
  () => controller.abort(new DOMException("Timed out waiting for the server to respond", "TimeoutError")),
  headersTimeoutMs,
)
return base(new Request(request, { signal: controller.signal })).finally(() => {
  clearTimeout(timer)
  release(entry)
})

Touches the same file as #47564; the conflict is confined to the test setup helper and the fetch body.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.