🐛 fixed static batch warmup

[joerunde]

Description

This changes the warmup for static batching back to how it originally was, only warming up a single pass.

This fixes a bug where the compiled model graphs were incorrect- we would invoke the model with one batch size and the model would output a different batch size.

[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 🚀