fix(supervisor): make agent__check a pure status probe that never consumes the handle

agent__check on a finished agent delegated to agent__collect, which
returned the full (unbounded) result and consumed the handle. That
contradicted the tool's own docs and broke the check-then-collect
pattern: a second collect on the same id failed.

check now reports { status: finished } with a pointer to
agent__collect and leaves the handle registered; collect is the single
retrieval verb. The tool description and prompt table are updated to
stop promising that check returns the result.

src/config/prompts.rs

@@ -82,7 +82,7 @@ pub(in crate::config) const DEFAULT_SPAWN_INSTRUCTIONS: &str = indoc! {"
     | Tool | Purpose |
     |------|----------|
     | `agent__spawn` | Spawn a subagent in the background. Returns an `id` immediately. |
-    | `agent__check` | Non-blocking check: is the agent done yet? Returns PENDING or result. |
+    | `agent__check` | Non-blocking status probe: running or finished. Never returns/consumes the result — use `agent__collect`. |
     | `agent__collect` | Blocking wait: wait for an agent to finish, return its output. |
     | `agent__list_available` | List all agent types you can spawn (name + description). Use this to discover specialists before calling `agent__spawn`. |
     | `agent__list_running` | List all subagents YOU have spawned, with their status. |