[BUG] Desktop: conversation freezes mid-"thinking", no error, must abort and resend — missing watchdog/retry UX

Description

On the Windows desktop app, quite often — especially with models that do long/reasoning "thinking" (e.g. OpenAI/DeepSeek-style reasoning dumps) — the conversation freezes mid-thinking: the agent is stuck in "thinking" state, spinner/tool call never resolves, and no output arrives.

The only workaround is to abort/terminate the conversation and resend the message (losing the reasoning context), which is disruptive for long debugging tasks.

Details of the bug

1. User sends a prompt (sometimes after long thinking, sometimes right at start).
2. Client shows "thinking…" (or the thinking block dumps) indefinitely; the stream silently stops; no error toast.
3. UI (desktop) is stuck; user must hit the abort/stop button and resent the exact same prompt; then it usually works — pointing to a transient stream/connection race, not the model being slow.

Associated issues (other symptoms, same "thinking stuck" family)

• #36370 (Desktop stuck in "thinking" - sidecar stream never completes, Windows)
• #37609 (Plan/Build mode sometimes hangs after long reasoning, never emits final response)
• #38951 (API 404 silently hangs in thinking state, no error recovery)
• #32366 (UI stuck on thinking indefinitely after stream error, no error displayed)
• #38644 (provider 500 drops response without showing error)
• #41069 (previously closed as duplicate of #37617 — free-tier model network blips treated as normal; user on this session reported similar)

Suggested improvements

1. Watchdog / timeout on the thinking phase: if no chunks received for N seconds (e.g. 60s), surface an explicit error/info banner ("model hasn't responded; retry?") and offer "retry with cached thinking".
2. Recover stuck turn automatically: detect stuck state and prompt user "Retry / discard and re-ask" instead of silently dangling.
3. Show provider error in the UI (many of these are provider 5xx/empty chunk cases): single toast for "provider error, auto-retried".
4. Non-lossy retry: when user aborts+resends, keep the previous reasoning/tool output as context so the retry isn't blind.

This is a frequent occurrence (dozens of similar reports) and the UX cost is high for long sessions.