feat(vision): 实现 mingxi-vision 推理服务核心功能

- engine/defect_classes: HRIPCB 6类缺陷定义 + 4类扩展 + severity排序
- engine/base: DefectBox / DetectResult 数据类 + BaseInferenceEngine 抽象基类
- engine/ultralytics_adapter: YOLOv8 .pt 推理适配器 + warmup
- engine/annotator: 按 severity 着色的标注图生成 + base64 编码
- engine/loader: 单例管理 + asyncio.Lock 推理串行化 + 热重载支持
- config: Pydantic v1 BaseSettings，读取 .env
- schema: DetectResponse / HealthResponse / ReloadRequest 等响应模型
- api/detect: POST /api/detect，内存读图，推理锁保护
- api/health: GET /api/health
- api/model: POST /api/model/reload 热重载接口
- main: FastAPI lifespan 启动加载模型

测试通过：/api/health 200，/api/detect 空图返回 pass，标注图 base64 正常

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

mingxi-vision/api/model.py

@@ -0,0 +1,16 @@
+from fastapi import APIRouter, HTTPException
+
+from engine.loader import get_inference_lock, reload_engine
+from schema import ReloadRequest, ReloadResponse
+
+router = APIRouter()
+
+
+@router.post("/model/reload", response_model=ReloadResponse)
+async def model_reload(req: ReloadRequest):
+    async with get_inference_lock():
+        try:
+            engine = reload_engine(req.model_path, req.runtime)
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=f"模型重载失败: {e}")
+    return ReloadResponse(status="ok", model_version=engine.model_version)