fix(ci): Python SDK CI failures — missing dotenv dep + sqlite3.Row AttributeError

[Copilot]

📋 Description

Python SDK CI job was failing at import time due to a missing python-dotenv dependency, and after unblocking that, all three omnibus tests failed with AttributeError: 'sqlite3.Row' object has no attribute 'get' in hsg.py.

Three fixes:

• pyproject.toml — added python-dotenv>=1.0 to dependencies. config.py calls from dotenv import load_dotenv at module level but the package was never declared.

• connectors/langchain.py — except ImportError stub only defined BaseChatMessageHistory and BaseRetriever; left BaseMessage, HumanMessage, AIMessage, Document, and CallbackManagerForRetrieverRun undefined. Any import of the module without langchain_core installed raised NameError at class definition time.

  # Before
  except ImportError:
      BaseChatMessageHistory = object
      BaseRetriever = object
  
  # After
  except ImportError:
      BaseChatMessageHistory = object
      BaseRetriever = object
      BaseMessage = object
      HumanMessage = object
      AIMessage = object
      Document = object
      CallbackManagerForRetrieverRun = object

• memory/hsg.py — compute_tag_match_score — function typed its mem param as dict but receives sqlite3.Row, which has no .get(). Replaced with direct bracket access guarded by try/except (KeyError, IndexError).

  # Before
  def compute_tag_match_score(mem: dict, q_toks: Set[str]) -> float:
      if not mem or not mem.get("tags"): return 0.0
  
  # After
  def compute_tag_match_score(mem, q_toks: Set[str]) -> float:
      if not mem: return 0.0
      try:
          tags_val = mem["tags"]
      except (KeyError, IndexError):
          return 0.0
      if not tags_val: return 0.0

🔄 Type of Change

• 🐛 Bug fix (non-breaking change which fixes an issue)
• ✨ New feature (non-breaking change which adds functionality)
• 💥 Breaking change (fix or feature that would cause existing functionality to not work as expected)
• 📚 Documentation update
• 🎨 Style/UI changes
• ♻️ Code refactoring
• ⚡ Performance improvements
• 🧪 Test updates
• 🔧 Build/CI changes

🧪 Testing

• I have tested this change locally
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

All 3 Python omnibus tests (test_evolutionary_stability, test_boolean_metadata_logic, test_content_robustness) pass. Node.js SDK job was already green and is unaffected.

📱 Screenshots (if applicable)

N/A

🔍 Code Review Checklist

• Code follows the project's coding standards
• Self-review of the code has been performed
• Code is properly commented, particularly in hard-to-understand areas
• Changes generate no new warnings
• Any dependent changes have been merged and published

📚 Related Issues

🚀 Deployment Notes

No deployment changes required.

📋 Additional Context

The sqlite3.Row bug would also silently swallow tag-match scoring in production (all tag scores returning 0.0) for any query path that reached compute_tag_match_score, so this fix has correctness impact beyond CI.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.