update and expand online example to continious batching

[yannicks1]

update openai_spyre_inference.py

changes:

• use model 'ibm-granite/granite-3.3-8b-instruct'
• use tp 4
• remove block-size argument as this gets overridden anyway (actually 2048 throws an error as not supported upstream)
• expand (and explain) server startup instructions for continuous batching

[github-actions]

👋 Hi! Thank you for contributing to vLLM support on Spyre.
Just a reminder: Make sure that your code passes all the linting checks, otherwise your PR won't be able to be merged. To do so, first install the linting requirements, then run format.sh and commit the changes. This can be done with uv directly:

uv sync --frozen --group lint --active --inexact

Or this can be done with pip:

uv pip compile --group lint > requirements-lint.txt
pip install -r requirements-lint.txt
bash format.sh

Now you are good to go 🚀

[joerunde]

should we use vllm serve here instead? Or python3 -m vllm.entrypoints.cli.main serve?

[yannicks1]

done

[maxdebayser]

Nice!