Deepen query dispatch seam with Command Topology Module (#3078)

* Deepen query dispatch seam with command topology module

* Stabilize SDK parity defaults and integration test gating

* docs(architecture): record pre-project config policy and e2e gate

* refactor(query): stop injecting native adapter in CLI dispatch path

* fix(config): align workflow auto-chain typing and docs

sdk/src/query/config-query.ts

@@ -16,6 +16,7 @@
  * ```
  */
 
+import { existsSync } from 'node:fs';
 import { readFile } from 'node:fs/promises';
 import { GSDError, ErrorClassification } from '../errors.js';
 import { loadConfig } from '../config.js';
@@ -174,6 +175,8 @@ export const resolveModel: QueryHandler = async (args, projectDir, workstream) =
     throw new GSDError('agent-type required', ErrorClassification.Validation);
   }
 
+  const configFilePath = planningPaths(projectDir, workstream).config;
+  const configExists = existsSync(configFilePath);
   const config = await loadConfig(projectDir, workstream);
   const profile = String(config.model_profile || 'balanced').toLowerCase();
 
@@ -188,9 +191,9 @@ export const resolveModel: QueryHandler = async (args, projectDir, workstream) =
     return { data: result };
   }
 
-  // resolve_model_ids: "omit" -- return empty string
+  // No project config (or explicit omit policy) -> return empty model id (CJS parity)
   const resolveModelIds = (config as Record<string, unknown>).resolve_model_ids;
-  if (resolveModelIds === 'omit') {
+  if (!configExists || resolveModelIds === 'omit') {
     const agentModels = MODEL_PROFILES[agentType];
     const result = agentModels
       ? { model: '', profile }