fix(model-fetch): 提取获取计划请求，并使网关运行时与共享构建器保持一致

- 将 build_execution_plan 的散参数收敛为 ModelFetchExecutionPlanRequest，消除 clippy too_many_arguments
- 在 aether-gateway 的 model_fetch runtime 中显式依赖 build_models_fetch_execution_plan
- 补充共享 models fetch plan builder 的运行时测试覆盖

apps/aether-gateway/src/model_fetch/runtime.rs

@@ -365,7 +365,9 @@ mod tests {
     use aether_data_contracts::repository::provider_catalog::{
         StoredProviderCatalogEndpoint, StoredProviderCatalogKey, StoredProviderCatalogProvider,
     };
-    use aether_model_fetch::{ModelFetchAssociationStore, ModelFetchTransportRuntime};
+    use aether_model_fetch::{
+        build_models_fetch_execution_plan, ModelFetchAssociationStore, ModelFetchTransportRuntime,
+    };
     use async_trait::async_trait;
     use serde_json::{json, Value};
     use std::collections::{HashMap, VecDeque};
@@ -746,6 +748,30 @@ mod tests {
         }
     }
 
+    #[tokio::test]
+    async fn gateway_runtime_state_supports_shared_models_fetch_plan_builder() {
+        let state = TestState::default();
+        let transport = sample_transport(
+            "openai",
+            "provider-openai",
+            "endpoint-openai-chat",
+            "key-openai-chat",
+            "openai:chat",
+            "api_key",
+            None,
+        );
+
+        let plan = build_models_fetch_execution_plan(&state, &transport)
+            .await
+            .expect("shared models fetch plan should build");
+
+        assert_eq!(plan.method, "GET");
+        assert_eq!(plan.provider_id, "provider-openai");
+        assert_eq!(plan.endpoint_id, "endpoint-openai-chat");
+        assert_eq!(plan.key_id, "key-openai-chat");
+        assert_eq!(plan.model_name.as_deref(), Some("models"));
+    }
+
     #[tokio::test]
     async fn model_fetch_uses_preset_models_without_endpoint() {
         let provider = sample_provider("provider-codex", "codex");