Enhance Markdown generation and external content control

- Integrate customized html2text library for flexible Markdown output
- Add options to exclude external links and images
- Improve content scraping efficiency and error handling
- Update AsyncPlaywrightCrawlerStrategy for faster closing
- Enhance CosineStrategy with generic embedding model loading

crawl4ai/extraction_strategy.py

@@ -68,7 +68,7 @@ class LLMExtractionStrategy(ExtractionStrategy):
         """
         super().__init__() 
         self.provider = provider
-        self.api_token = api_token or PROVIDER_MODELS.get(provider, None) or os.getenv("OPENAI_API_KEY")
+        self.api_token = api_token or PROVIDER_MODELS.get(provider, "no-token") or os.getenv("OPENAI_API_KEY")
         self.instruction = instruction
         self.extract_type = extraction_type
         self.schema = schema