get the token_type_ids from pooling params

[maxdebayser]

Description

The token type ids in upstream vllm are passed to the model runner in the pooling params. This PR fixes the preparation of token type inputs and also fixes the tensor sizes and dimensions to work both with the transformer modeling code and the vllm pooler code.

[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 🚀

[maxdebayser]

bot:test

[joerunde]

🌶️ , thanks for the extra test coverage too!