Feature Request: Support Query-Time ACL and RBAC enforcement in Azure AI Search

Azure AI Search now natively supports Microsoft Entra-based document-level security, which means we should be able to avoid hand-coded security trimming or complex role management.

- https://learn.microsoft.com/en-us/azure/search/search-document-level-access-overview
- https://learn.microsoft.com/en-au/azure/search/search-query-access-control-rbac-enforcement

Would it be possible to update the accelerator to handle AI Search endpoints that support this, via APIM policy that can pass-through end-user Entra ID tokens, in addition to current APIM (service-level) auth mechanisms?

Start by reading the linked Azure AI Search documentation on document-level security and query-time RBAC enforcement, then inspect the accelerator's APIM policy and service-level authentication flow. Determine how an end-user Entra ID token could pass through to supported AI Search endpoints while preserving current authentication. Done means the accelerator supports this configuration and documents or tests the resulting behavior.

Written by the indexing model from the issue text.