Add tests, server, resume training, and project cleanup

- Add 57 unit tests covering generators, models, and pipeline components
- Implement FastAPI HTTP service (server.py) with POST /solve and GET /health
- Add checkpoint resume (断点续训) to both CTC and regression training utils
- Fix device mismatch bug in CTC training (targets/input_lengths on GPU)
- Add pytest dev dependency to pyproject.toml
- Update .gitignore with data/solver/, data/real/, *.log
- Remove PyCharm template main.py
- Update training/__init__.py docs for solver training scripts

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

server.py

@@ -0,0 +1,88 @@
+"""
+FastAPI HTTP 推理服务
+
+提供 REST API 识别验证码:
+    POST /solve  - 上传图片，返回识别结果
+    GET  /health - 健康检查
+
+启动方式:
+    uv run python cli.py serve --port 8080
+
+请求示例:
+    curl -X POST http://localhost:8080/solve -F "image=@captcha.png"
+
+响应示例:
+    {"type": "normal", "result": "A3B8", "confidence": 0.95, "time_ms": 45}
+"""
+
+from __future__ import annotations
+
+
+def create_app():
+    """
+    创建 FastAPI 应用实例（工厂函数）。
+
+    cli.py 的 cmd_serve 依赖此签名。
+    需要安装 server 可选依赖: uv sync --extra server
+    """
+    from fastapi import FastAPI, File, Query, UploadFile
+    from fastapi.responses import JSONResponse
+
+    from inference.pipeline import CaptchaPipeline
+
+    app = FastAPI(
+        title="CaptchaBreaker",
+        description="验证码识别多模型系统 - HTTP 推理服务",
+        version="0.1.0",
+    )
+
+    pipeline: CaptchaPipeline | None = None
+
+    @app.on_event("startup")
+    def _load_models():
+        nonlocal pipeline
+        try:
+            pipeline = CaptchaPipeline()
+        except FileNotFoundError:
+            # 模型未导出时允许启动，但 /solve 会返回 503
+            pipeline = None
+
+    @app.get("/health")
+    def health():
+        models_loaded = pipeline is not None
+        return {"status": "ok" if models_loaded else "no_models", "models_loaded": models_loaded}
+
+    @app.post("/solve")
+    async def solve(
+        image: UploadFile = File(...),
+        type: str | None = Query(None, description="指定类型跳过分类"),
+    ):
+        if pipeline is None:
+            return JSONResponse(
+                status_code=503,
+                content={"error": "模型未加载，请先训练并导出 ONNX 模型"},
+            )
+
+        data = await image.read()
+        if not data:
+            return JSONResponse(
+                status_code=400,
+                content={"error": "空文件"},
+            )
+
+        try:
+            result = pipeline.solve(data, captcha_type=type)
+        except RuntimeError as e:
+            return JSONResponse(
+                status_code=400,
+                content={"error": str(e)},
+            )
+
+        return {
+            "type": result["type"],
+            "result": result["result"],
+            "raw": result["raw"],
+            "time_ms": result["time_ms"],
+        }
+
+    return app