docs: correct client rate-limit guidance

[sylvesterkaczmarek]

Summary

Correct the OpenAIClient and OpenAIClientAsync class-level documentation so it no longer claims that each client proactively handles API rate limiting.

Fixes #644.

Problem

Both client interfaces currently say:

The client also handles rate limiting per client. This means that creating and using multiple instances at the same time will not respect rate limits.

That implies client-wide proactive rate-limit coordination. The SDK does not implement that behavior. It can retry individual requests after retryable responses, including rate-limit responses, but it does not coordinate a shared rate budget across requests or across client instances.

The existing wording can therefore lead users to believe that reusing one client is required for correct API rate-limit accounting.

Fix

Update the shared guidance in both synchronous and asynchronous client interfaces to state that:

• reusing a client is still recommended for connection-pool/thread-pool efficiency;
• individual requests may be retried when the API returns a retryable response;
• the client does not proactively coordinate API rate limits across requests or client instances.

Scope

• OpenAIClient.kt: 3 additions / 3 deletions
• OpenAIClientAsync.kt: 3 additions / 3 deletions
• no API, runtime, retry, or transport behavior changes

Validation

The branch is based on current upstream main at 81134d727946fb84cc48db98e6c8c53c8a8d22ab and is 0 commits behind upstream.

Risk

Minimal. Documentation only, with sync and async client guidance kept consistent.