feat: Add delay/retry between requests for .extract

I have been trying to test using examples, but i keep being rate-limited so can there be an optional retry or delay between requests

```
langextract.inference.InferenceOutputError: Parallel inference error: Gemini API error: 429 RESOURCE_EXHAUSTED. {'error': {'code': 429, 'message': 'You exceeded your current quota, please check your plan and billing details. For more information on this error, head to: https://ai.google.dev/gemini-api/docs/rate-limits.', 'status': 'RESOURCE_EXHAUSTED', 'details': [{'@type': 'type.googleapis.com/google.rpc.QuotaFailure', 'violations': [{'quotaMetric': 'generativelanguage.googleapis.com/generate_content_free_tier_requests', 'quotaId': 'GenerateRequestsPerMinutePerProjectPerModel-FreeTier', 'quotaDimensions': {'location': 'global', 'model': 'gemini-2.5-flash'}, 'quotaValue': '10'}]}, {'@type': 'type.googleapis.com/google.rpc.Help', 'links': [{'description': 'Learn more about Gemini API quotas', 'url': 'https://ai.google.dev/gemini-api/docs/rate-limits'}]}, {'@type': 'type.googleapis.com/google.rpc.RetryInfo', 'retryDelay': '15s'}]}}
```

Start at the .extract entry point and trace how Gemini API requests are issued through the inference path. Review the 429 RESOURCE_EXHAUSTED response and its retryDelay, then define how an optional delay or retry setting should affect requests; done means rate-limited examples can recover or wait without breaking existing extraction behavior.

Written by the indexing model from the issue text.