[Bug]: Model UI shows obscure errors until the model is ready

[r4victor]

Steps to reproduce

1. Start a model service:

type: service
name: llama31

# If `image` is not specified, dstack uses its default image
python: 3.12
env:
  - HF_TOKEN
  - MODEL_ID=meta-llama/Meta-Llama-3.1-8B-Instruct
  - MAX_MODEL_LEN=4096
commands:
  - uv pip install vllm
  - vllm serve $MODEL_ID
    --max-model-len $MAX_MODEL_LEN
    --tensor-parallel-size $DSTACK_GPUS_NUM
port: 8000
# (Optional) Register the model
model: meta-llama/Meta-Llama-3.1-8B-Instruct

# Uncomment to leverage spot instances
#spot_policy: auto

resources:
  gpu: 24GB

2. Before the model is up and running, go to Model UI and try to chat. You'll see errors like "Connection errors" and CORS.

Actual behaviour

No response

Expected behaviour

Disable chat UI until the model is ready (if we know this due to probes).

dstack version

master

Server logs

Additional information

No response