multi-tenant Phase 1: user registry + per-user memory

- CDK: add agent-claw-users DynamoDB table (actor_id PK, RETAIN policy)
- CDK: grant agent-runner read/write on users table; add USERS_TABLE_NAME env
- CDK: fix cdk.json app field (was object, must be command string)
- CDK: add UsersTableName output
- agent-runner: get_or_create_user() auto-registers users on first contact
  (stores display_name, telegram_username, created_at, allowed)
- agent-runner: pass user_profile in AgentCore payload
- prompt_builder: split base prompt (cached) from per-user context (injected per-call)
  removes USER.md/MEMORY.md from shared load; user name/username injected dynamically
- main.py: extract user_profile from payload, build user_context string for prompt

agentclaw/app/agent_claw_main/main.py

@@ -140,8 +140,16 @@ def main(payload: dict, context) -> dict:
         region_name='us-east-1',
     )
 
-    # Build system prompt (cached across warm invocations)
-    system_prompt = build_system_prompt()
+    # Build system prompt — base cached, user context injected per-invocation
+    user_profile = payload.get('user_profile', {})
+    user_context = ''
+    if user_profile:
+        name = user_profile.get('display_name', '')
+        username = user_profile.get('telegram_username', '')
+        user_context = f'Name: {name}'
+        if username:
+            user_context += f'\nTelegram username: @{username}'
+    system_prompt = build_system_prompt(user_context=user_context)
 
     # Model: claude-sonnet-4-6 via cross-region inference
     model = BedrockModel(