Add debug logging for LLM query and response pipeline

[Copilot]

LLM responses received from API but not displaying in UI. Unclear if issue is in request construction, response parsing, or choices array access.

Changes

• Query logging: Log full query including topic context and neighboring topics

  console.debug('[LLM] Query with context:', { topicContext, userMessage, fullMessage })

• Response logging: Log complete API responses from OpenAI and Gemini before parsing

  console.debug('[LLM] OpenAI API response:', response.data)
  // Enables verification of response.data.choices[0].message.content structure

• Extraction logging: Log the actual message extracted from API response

  console.debug('[LLM] Extracted assistant message:', assistantMessage)

• Error context: Log full response object when choices/candidates array missing

Debugging

Logs trace the full pipeline: query construction → API call → response parsing → message extraction. Opens DevTools console and filter for [LLM] to identify failure point.

Updated LLM_INTEGRATION.md with debugging workflow and expected log format.

---

💡 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.