fix: implement default stepCountIs(5) stop condition and fix flaky test

- Added missing default stop condition in shouldStopExecution() that was
  documented but not implemented, preventing infinite tool execution loops
- Changed test model from llama-3.1-8b-instruct to claude-sonnet-4.5 to
  fix timeout issues in chat-style tools test

Author: mattapperson

src/lib/model-result.ts

@@ -47,7 +47,7 @@ import {
 import { executeTool } from './tool-executor.js';
 import { executeNextTurnParamsFunctions, applyNextTurnParamsToRequest } from './next-turn-params.js';
 import { hasExecuteFunction } from './tool-types.js';
-import { isStopConditionMet } from './stop-conditions.js';
+import { isStopConditionMet, stepCountIs } from './stop-conditions.js';
 
 /**
  * Type guard for stream event with toReadableStream method
@@ -284,13 +284,15 @@ export class ModelResult<TTools extends readonly Tool[]> {
   /**
    * Check if stop conditions are met
    * Returns true if execution should stop
+   * Default: stepCountIs(5) if no stopWhen is specified
    */
   private async shouldStopExecution(): Promise<boolean> {
-    if (!this.options.stopWhen) return false;
+    // Use default of stepCountIs(5) if no stopWhen is specified
+    const stopWhen = this.options.stopWhen ?? stepCountIs(5);
 
-    const stopConditions = Array.isArray(this.options.stopWhen)
-      ? this.options.stopWhen
-      : [this.options.stopWhen];
+    const stopConditions = Array.isArray(stopWhen)
+      ? stopWhen
+      : [stopWhen];
 
     return isStopConditionMet({
       stopConditions,

tests/e2e/call-model.test.ts

@@ -160,7 +160,7 @@ describe('callModel E2E Tests', () => {
 
     it('should work with chat-style messages and chat-style tools together', async () => {
       const response = client.callModel({
-        model: 'meta-llama/llama-3.1-8b-instruct',
+        model: 'anthropic/claude-sonnet-4.5',
         input: fromChatMessages([
           {
             role: 'system',