feat(model): 新增 ModelLoader 链式构建 API 及 ORT GPU/Tract 多线程配置

- 在 ddddocr-core 中定义 ModelBuilder Trait 及其错误类型
- ddddocr-ort 支持 use_gpu、device_id 及 num_threads 链式配置与 CUDA 硬件加速
- ddddocr-tract 基于 multithread-mm 特性支持 CPU 线程数控制
- 支持基于 tract-linalg 配置推理线程数，显式引入 tract-linalg 的 multithread-mm 特性，控制 GEMM 算子并发
- 优化线程池加载策略，适配 Tokio 异步及 CLI 等多场景

ddddocr-core/Cargo.toml

@@ -5,12 +5,12 @@ edition = { workspace = true }
 license = { workspace = true }
 
 [dependencies]
-image = "0.25.10"
-base64 = "0.22.1"
-imageproc = { version = "0.26.2", default-features = true }
-serde = { workspace = true }
-serde_json = "1.0.150"
 ndarray = { workspace = true } # 继承自工作空间
+
+base64 = { workspace = true }
+
+image = { workspace = true }
+imageproc = { workspace = true }
+
 thiserror = { workspace = true } # 刚好可以开始接入你需要的标准库错误处理
-tracing={workspace = true}
-#serde = { workspace = true, features = ["derive"] }
+tracing={workspace = true}