chore: auto load model on /chat/completions request

[louis-jan]

Describe Your Changes

This PR is to add model loading on the /chat/completions & /embeddings request from the JS layer.

Note: We use cortexso/nomic-embed-text-v1 as the default embedding model, in case the model is not defined. It will automatically pull embedding model if it does not exit.

CLI update should be covered by refactoring to send REST requests.

sequenceDiagram
    participant User
    participant API_Controller
    participant Engine_Server
    participant Model

    User->>API_Controller: Chat request
    API_Controller->>Engine_Server: Check Model Running
    Engine_Server-->>API_Controller: Model status
    alt Model is not running
        API_Controller->>Engine_Server: Start Model command
        Engine_Server->>Model: Start Model
        Model-->>Engine_Server: Model started
        Engine_Server-->>API_Controller: Model ready
    end
    API_Controller->>Model: Send Chat Completions
    Model-->>API_Controller: Return Completions
    API_Controller-->>User: Send Response

Loading

Fixes Issues

• Closes #

Self Checklist

• Added relevant comments, esp in complex areas
• Updated docs (for bug fixes / features)
• Created issues for follow-up changes or refactoring needed