feat(chat): scaffold chatbotAPI with local Ollama integration

[Yatharth-coding]

Executive Summary

This PR introduces the foundational backend architecture for the AI Circuit Assistant (Minor Project). It scaffolds the new chatbotAPI Django app and integrates a secure, fully local LLM client using Ollama. This ensures the AI assistant can run entirely locally without relying on paid external APIs, API keys, or internet connectivity.

Key Features & Implementations

• chatbotAPI Scaffold: Created the core Django app structure following the project's existing API conventions, including URL routing and app registration.
• Local Ollama Integration:
  • Built services/llm_client.py to handle HTTP communication with a local Ollama instance.
  • Implemented robust error handling for timeouts, connection drops, and missing models.
• Core Endpoints:
  • GET /api/chat/health/: Added an open health check endpoint.
  • POST /api/chat/message/: Implemented the main chat interface featuring strict payload validation (enforcing an 8KB limit on context JSON to prevent oversized requests).
• Automated Testing: Developed comprehensive unit tests utilizing unittest.mock to ensure that CI/CD pipelines will pass without requiring a live Ollama server to be running.

Local Testing Instructions for Reviewers

1. Install Ollama on your host machine.
2. Pull the required model by running: ollama pull llama3.2 (or the model specified in your .env).
3. Ensure Ollama is running locally on port 11434.
4. Start the eSim-Cloud Docker stack and test the /api/chat/message/ endpoint.