bug: `cortex chat` is a non-interactive request & should stdout to terminal

[freelerobot]

Cortex version

v83

Describe the Bug

Right now cortex chat starts an interactive shell.
Originally, we only intended for run to start the shell.

For consistency, cortex chat should mirror the other cortex subcommands like models, engines: ping the api endpoint, and return a response within the standard terminal.

But cmiiw @dan-homebrew

Steps to Reproduce

1. cortex start the server
2. cortex models start ... the model
3. cortex chat ...
4. expected: chat request response within terminal

Screenshots / Logs

Instead, we enter an interactive chat shell.

(base) PS C:\Users\n\cortexcpp-nightly\models> cortex-nightly chat mistral -m "hello"
Inorder to exit, type `exit()`
> Hello! How can I assist you today?
>

What is your OS?

• MacOS
• Windows
• Linux

What engine are you running?

• cortex.llamacpp (default)
• cortex.tensorrt-llm (Nvidia GPUs)
• cortex.onnx (NPUs, DirectML)

[freelerobot]

works wonderfully