Support triggering prompt cache writes in the orchestration service

[christophmeyer]

Checklist

• My issue is not an AI Core service issue, but an SDK issue.
• My issue is a Python related issue, not a JS/TS or Java related issue.
• My issue is not already reported or documented.

Describe the Problem

The orchestration service client (v1 gen_ai_hub/orchestration/ and v2 gen_ai_hub/orchestration_v2/) has no way to trigger/create prompt-cache writes.

Propose a Solution

Requesting support for cache-write semantics (e.g. cache_control / cache-point markers on messages or config).

Describe Alternatives

No response

Affected Development Phase

Development

Impact

Impaired

Timeline

No response

Additional Context

No response

[yamaceay]

Hi @christophmeyer , does the current PR resolve your issue?