Get the org-level llm_judge model default

The org-level default model for llm_judge policy checks.

model_settings is the org's chosen ModelChainEntry ({model, config}), or null when unset — in which case the runtime falls back to the code role default (gpt-5.4-mini / effort low). The UI renders that fallback label from the model catalog's policy_judge role default, so this response carries ONLY the override.