fix: add 'usage limit' to rate limit detection patterns

OpenAI/ChatGPT returns "You have hit your ChatGPT usage limit (plus plan)"
when users exceed their plan quota. This error wasn't being recognized as a
rate limit, so fallback to alternative models wasn't triggering.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/agents/pi-embedded-helpers.test.ts

@@ -272,6 +272,14 @@ describe("classifyFailoverReason", () => {
     );
     expect(classifyFailoverReason("bad request")).toBeNull();
   });
+
+  it("classifies OpenAI usage limit errors as rate_limit", () => {
+    expect(
+      classifyFailoverReason(
+        "You have hit your ChatGPT usage limit (plus plan)",
+      ),
+    ).toBe("rate_limit");
+  });
 });
 
 describe("isCloudCodeAssistFormatError", () => {

src/agents/pi-embedded-helpers.ts

@@ -343,6 +343,7 @@ const ERROR_PATTERNS = {
     "resource has been exhausted",
     "quota exceeded",
     "resource_exhausted",
+    "usage limit",
   ],
   timeout: [
     "timeout",