fix: #477 Add defensive error handling for malformed LLM response choices

[suhaniiz]

📋 PR Checklist

Thank you for contributing to PDF-Assistant-RAG! 🎉
Please fill out this template before submitting. PRs without it filled in will be closed.

---

🔗 Related Issue

Closes #477

---

📝 What does this PR do?

Adds defensive attribute retrieval using getattr when parsing the LLM client's chat completion choices. Previously, the logic assumed that if response.choices was populated, the nested .message and .content attributes were guaranteed to exist, causing an unhandled AttributeError on unexpected schemas or partial API responses. This fix ensures the application safely falls back to returning None without crashing.

---

🗂️ Type of Change

• 🐛 Bug fix
• ✨ New feature
• 🔧 Refactor / code cleanup
• 📝 Documentation update
• 🎨 UI / styling change
• ⚙️ CI / tooling / config change
• 🧪 Tests

---

🧪 How was this tested?

• Tested the affected API endpoints manually
• Verified that the code changes cleanly replace only the target fallback block without touching unrelated application logic.

---

📸 Screenshots (if UI change)

N/A - Backend logic fix only.

---

⚠️ Anything to flag for reviewers?

No major concerns. The change is isolated strictly inside generate_document_summary and builds upon the existing default return patterns.

---

✅ Self-Review Checklist

• My branch is based on dev, not main
• I have not added any secrets / API keys
• I have not modified main branch or any HuggingFace deployment config
• My code follows the existing style (no unnecessary formatting changes)
• I have updated relevant docs / comments if needed

[suhaniiz]

@param20h , this pr is under gssoc 2026, kindly review it