fix: guard against None values in _apply_transform

[jnMetaCode]

Summary

JsonCssExtractionStrategy._apply_transform() crashes with AttributeError when value is None. This happens when a CSS or XPath selector finds no matching element — the extracted value is None, and calling .lower(), .upper(), or .strip() on it raises:

AttributeError: 'NoneType' object has no attribute 'lower'

Changes

• extraction_strategy.py: Added an early return guard at the start of _apply_transform:

  if value is None:
      return value

Impact

Without this fix, any extraction schema with transforms will crash on pages where optional fields are missing. This is a common scenario when scraping heterogeneous pages. The fix allows transforms to gracefully pass through None values.