feat: AI assistant phase 2 - configurable prompt, action operations, FAQ matching, NVIDIA provider

- Admin-configurable AI prompt/quick questions from system_configs DB
- GET /api/v1/ai/quick-questions endpoint for fetching quick questions
- Local FAQ matching for instant responses (avoid AI calls for common Qs)
- AI action extraction: "add customer" intent detected, structured data returned
- Frontend action confirmation card with editable fields, calls customer API on confirm
- NVIDIA provider (stepfun-ai/step-3.5-flash) for faster chat vs deepseek-v4-flash
- Fixed httpx client timeout preventing backend hangs
- Added log_usage calls for auth events (register/login/guest/wechat)
- Admin tabs (users/stats/logs/config) fully functional with real backend
- AiAssistant component added to all tabbar pages

backend/app/ai/providers/__init__.py

@@ -5,5 +5,6 @@ from .local import LocalProvider
 from .spark import SparkProvider
 from .sensenova import SensenovaProvider
 from .opencode_go import OpencodeGoProvider
+from .nvidia import NvidiaProvider
 
-__all__ = ["OpenAIProvider", "ClaudeProvider", "DeepLProvider", "LocalProvider", "SparkProvider", "SensenovaProvider", "OpencodeGoProvider"]
+__all__ = ["OpenAIProvider", "ClaudeProvider", "DeepLProvider", "LocalProvider", "SparkProvider", "SensenovaProvider", "OpencodeGoProvider", "NvidiaProvider"]

backend/app/ai/providers/nvidia.py

@@ -0,0 +1,50 @@
+from typing import Dict, Any, Optional, List
+from app.ai.providers.openai import OpenAIProvider, SYSTEM_PROMPTS
+import logging
+import time
+import httpx
+
+logger = logging.getLogger(__name__)
+
+
+class NvidiaProvider(OpenAIProvider):
+    def __init__(self, api_key: str, model: str = "stepfun-ai/step-3.5-flash", base_url: str = "https://integrate.api.nvidia.com/v1"):
+        super().__init__(
+            api_key=api_key,
+            model=model,
+            base_url=base_url,
+            http_client=httpx.AsyncClient(timeout=httpx.Timeout(60.0)),
+        )
+        self._name = f"nvidia-{model}"
+
+    async def chat(self, message: str, history: list = None, system_prompt: str = None) -> Dict[str, Any]:
+        t0 = time.time()
+
+        system = system_prompt or SYSTEM_PROMPTS["chat"]
+        messages = [{"role": "system", "content": system}]
+        if history:
+            for h in history[-10:]:
+                messages.append(h)
+        messages.append({"role": "user", "content": message})
+        t1 = time.time()
+
+        kwargs = {
+            "model": self.model,
+            "messages": messages,
+            "max_tokens": 300,
+            "temperature": 0.3,
+        }
+        resp = await self.client.chat.completions.create(**kwargs)
+        t2 = time.time()
+
+        content = resp.choices[0].message.content or ""
+        if not content and hasattr(resp.choices[0].message, "reasoning"):
+            content = resp.choices[0].message.reasoning
+        t3 = time.time()
+
+        logger.info(
+            f"NVIDIA timing: build_msgs={t1-t0:.1f}s api_call={t2-t1:.1f}s process={t3-t2:.1f}s "
+            f"chars_in={sum(len(m.get('content','')) for m in messages)} chars_out={len(content)}"
+        )
+
+        return {"reply": content, "provider": self.name, "model": self.model}

backend/app/ai/providers/openai.py

@@ -14,11 +14,21 @@ SYSTEM_PROMPTS = {
                  "marketing content that drives action. Adapt to the target audience's culture. "
                  "Return ONLY the copy, no explanations.",
     "extract": "You extract structured data from text. Return ONLY valid JSON matching the requested schema.",
+    "chat": "你是 TradeMate（外贸小助手）的 AI 助手。你的职责是帮助外贸从业者解答关于本工具使用的问题，以及提供外贸业务建议。\n"
+            "你可以回答的问题包括：\n"
+            "- 功能介绍：翻译、客户管理、产品管理、报价单、营销文案、WhatsApp 集成等\n"
+            "- 使用帮助：如何添加客户、如何生成报价单、如何导出数据等\n"
+            "- 外贸知识：贸易术语（FOB、CIF 等）、谈判技巧、跟进策略等\n\n"
+            "回答要求：\n"
+            "- 简洁扼要，用中文回答\n"
+            "- 涉及操作步骤时用数字列表说明\n"
+            "- 不确定的问题不要编造，直接说需要查证\n"
+            "- 语气友好专业",
 }
 
 
 class OpenAIProvider(AIProvider):
-    def __init__(self, api_key: str, model: str = "gpt-4o", base_url: Optional[str] = None):
+    def __init__(self, api_key: str, model: str = "gpt-4o", base_url: Optional[str] = None, http_client=None):
         try:
             from openai import AsyncOpenAI
         except ImportError:
@@ -29,6 +39,8 @@ class OpenAIProvider(AIProvider):
         kwargs = {"api_key": api_key}
         if base_url:
             kwargs["base_url"] = base_url
+        if http_client:
+            kwargs["http_client"] = http_client
         self.client = AsyncOpenAI(**kwargs)
         self.model = model
         self._name = f"openai-{model}"
@@ -92,6 +104,24 @@ class OpenAIProvider(AIProvider):
         except json.JSONDecodeError:
             return {"data": {}, "confidence": 0.0, "provider": self.name, "error": "parse_failed"}
 
+    async def chat(self, message: str, history: list = None, system_prompt: str = None) -> Dict[str, Any]:
+        system = system_prompt or SYSTEM_PROMPTS["chat"]
+        messages = [{"role": "system", "content": system}]
+        if history:
+            for h in history[-10:]:
+                messages.append(h)
+        messages.append({"role": "user", "content": message})
+
+        kwargs = {
+            "model": self._cheap_model,
+            "messages": messages,
+            "max_tokens": 2000,
+            "temperature": 0.7,
+        }
+        resp = await self.client.chat.completions.create(**kwargs)
+        content = resp.choices[0].message.content or ""
+        return {"reply": content, "provider": self.name, "model": self.model}
+
     async def _call(self, system: str, prompt: str, max_tokens: int = 3000, response_format: Optional[Dict] = None, model: Optional[str] = None) -> str:
         kwargs = {
             "model": model or self.model,