Feature Request: Add built-in retry logic for runner creation

[xgboosted]

It would be very helpful if the Cyclenerd/hcloud-github-runner action supported built-in retry logic for the runner creation step. Sometimes, runner creation can fail due to resource unavailability at Hetzner Cloud, transient network or API issues, and having a configurable retry mechanism (e.g., number of attempts, delay between retries) would make CI pipelines more robust and reduce manual intervention.

Suggested enhancement:

• Add inputs to configure the number of retry attempts and delay between attempts.
• If runner creation fails, the action should automatically retry up to the configured limit before failing the step.

This would allow users to handle intermittent failures gracefully without having to implement custom retry logic in their workflows.

Thank you for considering this feature!