fix(runtime-fallback): abort stuck subagent on quota error with no fallback

When a subagent session (e.g. Momus on GPT) hits a quota/usage-limit
error and the agent's category has no `fallback_models` configured, the
runtime-fallback hook previously returned silently. The OpenCode session
stayed in `retry` status indefinitely while the SDK kept hitting the
limit, the sync-task poller treated `retry` as active work, and the
parent's pending `task` tool call never resolved — leaving a stuck
"waiting for subagent" indicator in the parent conversation.

Narrow fix: at the `fallbackModels.length === 0` exit point, if the
session is a known subagent AND the error classifies as
`quota_exceeded`, abort the subagent session. The existing
`getTerminalSessionError` path in `sync-session-poller.ts` then surfaces
the error via the parent's tool result, which is the persistent surface
the user is already watching.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/hooks/runtime-fallback/message-update-handler.ts

@@ -8,6 +8,7 @@ import { getFallbackModelsForSession } from "./fallback-models"
 import { resolveFallbackBootstrapModel } from "./fallback-bootstrap-model"
 import { dispatchFallbackRetry } from "./fallback-retry-dispatcher"
 import { hasVisibleAssistantResponse } from "./visible-assistant-response"
+import { subagentSessions } from "../../features/claude-code-session-state"
 
 export { hasVisibleAssistantResponse } from "./visible-assistant-response"
 
@@ -112,6 +113,16 @@ export function createMessageUpdateHandler(deps: HookDeps, helpers: AutoRetryHel
       const fallbackModels = getFallbackModelsForSession(sessionID, resolvedAgent, pluginConfig)
 
       if (fallbackModels.length === 0) {
+        if (
+          subagentSessions.has(sessionID) &&
+          classifyErrorType(error) === "quota_exceeded"
+        ) {
+          log(`[${HOOK_NAME}] Aborting subagent on unrecoverable quota error (no fallback configured)`, {
+            sessionID,
+            model,
+          })
+          await helpers.abortSessionRequest(sessionID, "message.updated.subagent-quota-no-fallback")
+        }
         return
       }