feat(intelligence): complete migration to Rust backend

- Unify all intelligence modules to use intelligenceClient
- Delete legacy TS implementations (agent-memory, reflection-engine, heartbeat-engine, context-compactor, agent-identity, memory-index)
- Update all consumers to use snake_case backend types
- Remove deprecated llm-integration.test.ts

This eliminates code duplication between frontend and backend, resolves
localStorage limitations, and enables persistent intelligence features.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

desktop/src/lib/skill-discovery.ts

@@ -11,7 +11,7 @@
  * Reference: ZCLAW_AGENT_INTELLIGENCE_EVOLUTION.md §6.5.2
  */
 
-import { getMemoryManager } from './agent-memory';
+import { intelligenceClient } from './intelligence-client';
 import { canAutoExecute } from './autonomy-manager';
 
 // === Types ===
@@ -295,8 +295,9 @@ export class SkillDiscoveryEngine {
 
       // 3. Check memory patterns for recurring needs
       try {
-        const memories = await getMemoryManager().search(skill.name, {
+        const memories = await intelligenceClient.memory.search({
           agentId,
+          query: skill.name,
           limit: 5,
           minImportance: 3,
         });