[FEATURE] Built-in Automatic Retry & Backoff Middleware

Confirmation of Request Source

• I confirm that this feature request is for the xAI SDK Python library (e.g., client features, SDK methods, or documentation) and not for the xAI API itself (e.g., model capabilities or API changes).

Describe the feature you'd like

Concept: Add customizable retry policies for gRPC rate limits (RESOURCE_EXHAUSTED) and transient network dropouts (UNAVAILABLE).

Why it matters: Network glitches and API rate limits currently throw unhandled gRPC errors. A built-in exponential backoff decorator makes production applications resilient out of the box.

Additional Context

Implementation Scope:

from xai_sdk import Client, RetryPolicy

client = Client(
    retry_policy=RetryPolicy(
        max_retries=3,
        backoff_factor=1.5,
        retry_on=[grpc.StatusCode.UNAVAILABLE, grpc.StatusCode.RESOURCE_EXHAUSTED]
    )
)

Start by locating the Python SDK's Client construction and current gRPC error handling. Review how UNAVAILABLE and RESOURCE_EXHAUSTED failures are surfaced, then define completion around configurable retry counts, exponential backoff, and status-code selection through the proposed RetryPolicy interface.