feat: Add Ollama support for local AI models

This commit introduces support for using Ollama to run large language models locally within AgentGPT. Key changes: - Modified backend settings and model factory to recognize and use Ollama as a model provider if configured via the `OLLAMA_API_BASE` environment variable. - Updated frontend components (model types, stores, constants) to list and allow selection of common Ollama models (llama2, mistral, codellama). - Enhanced the CLI setup script to prompt users for Ollama configuration (API base URL) and update the .env file accordingly. - Updated API endpoint for listing models to correctly reflect Ollama model availability based on server configuration. - Added documentation: a new page explaining Ollama setup and usage, and updated the main README.