feat(agent): Jina 搜索改用标准 MCP SDK 接入

- agent/task_executor.py: Jina 搜索从手搓 httpx 改为官方 mcp SDK
  (streamablehttp_client + ClientSession);工具经 OpenAI function-calling 暴露给模型
- agent/requirements.txt: +mcp==1.28.0;pydantic 2.9.2->2.13.4(mcp 要求 >=2.11)
- orchestrator/agent_launcher.py: JINA_API_KEY 经 per-swarm Secret 透传给 agent pod
  (SENSITIVE_ENV_KEYS),不内联 PodSpec
- k8s/orchestrator-local.yaml: 本地部署清单(默认 in-pod 沙箱评估开关 + JINA_API_KEY)

沙箱保持默认 in-pod 方案,未引入 OpenSandbox。
影响范围: agent_swarm(agent/orchestrator) + Agent(新增 Jina MCP 工具)。
密钥经 k8s Secret 注入无明文。不影响 Manager 契约/计费/审计/发布链路。

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
(cherry picked from commit deb984ac38)

k8s/orchestrator-local.yaml

@@ -87,6 +87,19 @@ spec:
                   name: swarm-model-key
                   key: OPENAI_API_KEY
                   optional: true
+            # Sandbox: default in-pod code-execution quality evaluation, fail-closed double gate
+            # (ENABLE_QUALITY_EVAL + HEICODE_SANDBOX_ISOLATED). Matches the deployed environment.
+            - name: ENABLE_QUALITY_EVAL
+              value: "1"
+            - name: HEICODE_SANDBOX_ISOLATED
+              value: "1"
+            # Jina MCP key — passed through to launched agent pods via agent_launcher
+            - name: JINA_API_KEY
+              valueFrom:
+                secretKeyRef:
+                  name: swarm-jina-key
+                  key: JINA_API_KEY
+                  optional: true
           resources:
             requests: { memory: "256Mi", cpu: "200m" }
             limits: { memory: "512Mi", cpu: "500m" }