[FEATURE REQ] Port hub region caching per partition level from .NET SDK

[jeet1995]

Summary

Port the per-partition hub region caching feature from .NET SDK (PR Azure/azure-cosmos-dotnet-v3#5648) to the Java SDK.

Background

On single-master accounts encountering repeated 404/1002 (ReadSessionNotAvailable) errors, the SDK should discover the hub region via a 403/3 (WriteForbidden) discovery chain and cache the result per partition. Subsequent requests to the same partition route directly to the cached hub, eliminating redundant discovery round-trips.

Key Behavior

• After 2 consecutive 404/1002 on a single-master account, set \x-ms-cosmos-hub-region-processing-only\ header
• Non-hub regions return 403/3 (WriteForbidden) — SDK retries to next region (discovery chain)
• Hub region responds with 200 OK — SDK caches hub URI for that partition
• Future requests route directly to cached hub (warm path)
• Works for both PPAF and non-PPAF accounts
• Must be gated by a feature flag / environment variable (per Debdatta's guidance)

.NET Reference

• PR: [Internal] PPAF: Adds Hub Region Caching Per Partition Level azure-cosmos-dotnet-v3#5648
• Key files: ClientRetryPolicy.cs, GlobalPartitionEndpointManagerCore.cs, GatewayStoreModel.cs

Java SDK Files Likely Affected

• \ClientRetryPolicy.java\
• \GlobalPartitionEndpointManager.java\ / related partition failover classes
• \RxGatewayStoreModel.java\
• \Configs.java\ (feature flag)

Acceptance Criteria

• Hub region discovery via 403/3 chain after 2x 404/1002 on single-master
• Per-partition hub region caching (warm path skips discovery)
• Feature flag gate (environment variable, disabled by default initially)
• Works for both PPAF and non-PPAF accounts
• Unit tests + integration tests