[Bug]: The maxtokens setting is too small and there is no correct prompt

[TimingZ]

crawl4ai version

0.7.6

Expected Behavior

using the LLMExtractionStrategy, model choice gemini

Current Behavior

When using the LLMExtractionStrategy, if the maxtokens are set too small, the context window is insufficient to carry the output content of the LLM, so there is no return value. However, the returned Json of the LLM, such as "finish_reason":"MAX_TOKENS", does not return to the location of the outermost method call, making it impossible for the caller to determine the core of the problem

Is this reproducible?

Yes

Inputs Causing the Bug

Steps to Reproduce

Code snippets

result = await crawler.arun(
            url="example.com", config=crawler_config
        )

print(result.extracted_content)
//output
[
    {
        "index": 0,
        "error": true,
        "tags": [
            "error"
        ],
        "content": "'NoneType' object has no attribute 'startswith'"
    }
]

OS

Windows

Python version

3.12.9

Browser

No response

Browser version

No response

Error logs & Screenshots (if applicable)

[
{
"index": 0,
"error": true,
"tags": [
"error"
],
"content": "'NoneType' object has no attribute 'startswith'"
}
]