初始化换发型项目：3个微服务代码 + 部署脚本

包含：
- hair_service_sd: 换发型/换发色算法服务 (端口 8801)
- photo_service: LoRA 训练调度服务 (端口 32678)
- stable-diffusion-webui: SD WebUI 推理服务 (端口 57860)
- kohya_ss_home: 训练环境代码
- meidaojia: 监控测试脚本
- setup.sh: 一键部署脚本 (conda环境恢复 + 配置生成 + 完整性检查)
- start_all_services.sh: 启动3个服务
- configure.ini.template: 路径模板化 (BASE_DIR自动推导)
- conda_envs/py310.yml: py310 环境定义

大文件 (weights/, models/, data/, conda_envs/*.tar.gz 等) 通过 .gitignore 排除，
由网盘单独上传。

stable-diffusion-webui/modules/textual_inversion/ui.py

@@ -0,0 +1,38 @@
+import html
+
+import gradio as gr
+
+import modules.textual_inversion.textual_inversion
+from modules import sd_hijack, shared
+
+
+def create_embedding(name, initialization_text, nvpt, overwrite_old):
+    filename = modules.textual_inversion.textual_inversion.create_embedding(name, nvpt, overwrite_old, init_text=initialization_text)
+
+    sd_hijack.model_hijack.embedding_db.load_textual_inversion_embeddings()
+
+    return gr.Dropdown.update(choices=sorted(sd_hijack.model_hijack.embedding_db.word_embeddings.keys())), f"Created: {filename}", ""
+
+
+def train_embedding(*args):
+
+    assert not shared.cmd_opts.lowvram, 'Training models with lowvram not possible'
+
+    apply_optimizations = shared.opts.training_xattention_optimizations
+    try:
+        if not apply_optimizations:
+            sd_hijack.undo_optimizations()
+
+        embedding, filename = modules.textual_inversion.textual_inversion.train_embedding(*args)
+
+        res = f"""
+Training {'interrupted' if shared.state.interrupted else 'finished'} at {embedding.step} steps.
+Embedding saved to {html.escape(filename)}
+"""
+        return res, ""
+    except Exception:
+        raise
+    finally:
+        if not apply_optimizations:
+            sd_hijack.apply_optimizations()
+