Reduce timeout for token inquery

What does this MR do and why?

Sometimes the external API call to obtain tokens is taking too long. Instead of the 30 seconds timeout we set a more reasonable timeout of 3 seconds.

This is happening from time to time:

• incident: https://gitlab.com/gitlab-com/gl-infra/production/-/work_items/21728
• the further investigation gitlab-org/modelops/applied-ml/code-suggestions/ai-assist#2273

References

Screenshots or screen recordings

Before	After

How to set up and validate locally

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.