feat: add privacy mode, thinking levels, PWA support, and i18n

- Add privacy mode to hide IP and user preferences from AI requests
- Add model thinking levels (low/medium/high) for context analysis depth
- Add PWA support with service worker, manifest, and app icons
- Add SettingsPanel for user preferences (theme, background, language)
- Add i18n translations for en/zh/ja/ko/de/fr
- Add Pinia store for centralized settings management
- Update backend to support user preferences and thinking levels
- Update config to use absolute API URLs

backend/.env.example

@@ -1,4 +1,4 @@
 OPENAI_API_KEY=ollama
 OLLAMA_BASE_URL=http://192.168.0.120:11434/v1/
-OLLAMA_MODEL=gpt-oss:120b
+OLLAMA_MODEL=gpt-oss:20b
 VLM_MODEL=qwen3-vl:30b

backend/llm.py

@@ -7,7 +7,7 @@ from dotenv import load_dotenv
 
 load_dotenv()
 
-OLLAMA_MODEL = os.getenv('OLLAMA_MODEL', 'gpt-oss:120b')
+OLLAMA_MODEL = os.getenv('OLLAMA_MODEL', 'gpt-oss:20b')
 OLLAMA_HOST = os.getenv('OLLAMA_HOST', 'http://192.168.0.120:11434')
 VLM_MODEL = os.getenv('VLM_MODEL', 'qwen3-vl:30b')
 

backend/main.py

@@ -27,10 +27,20 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+from typing import Optional
+
+class UserPreferences(BaseModel):
+    language: str = 'auto'
+    currency: str = 'auto'
+    timezone: str = 'auto'
+
 class CompletionRequest(BaseModel):
     prefix: str
     suffix: str
     languageId: str = 'markdown'
+    model_thinking: str = 'low'
+    privacy_mode: bool = False
+    user_preferences: Optional[UserPreferences] = None
 
 class OCRRequest(BaseModel):
     image: str
@@ -50,26 +60,40 @@ def get_client_ip(request: Request) -> str:
 @app.post("/v1/completions")
 async def create_completion(request: Request, req: CompletionRequest):
     request_id = str(uuid.uuid4())[:8]
-    client_ip = get_client_ip(request)
-    # 查询 IP 归属地
-    location = get_ip_location_text(client_ip)
-    if location:
-        logger.info("[%s] client_location=%s", request_id, location)
+    
+    client_ip = "hidden"
+    location = ""
+    
+    if not req.privacy_mode:
+        client_ip = get_client_ip(request)
+        # 查询 IP 归属地
+        location = get_ip_location_text(client_ip)
+        if location:
+            logger.info("[%s] client_location=%s", request_id, location)
+    
     try:
         logger.info(
-            "[%s] /v1/completions client_ip=%s prefix_chars=%d suffix_chars=%d lang=%s prefix_tail='%s' suffix_head='%s'",
+            "[%s] /v1/completions client_ip=%s prefix_chars=%d suffix_chars=%d lang=%s thinking=%s privacy=%s",
             request_id,
             client_ip,
             len(req.prefix or ""),
             len(req.suffix or ""),
             req.languageId,
-            _preview((req.prefix or "")[-120:]),
-            _preview((req.suffix or "")[:120]),
+            req.model_thinking,
+            req.privacy_mode
         )
         llm_prefix, llm_suffix = prepare_prompt_context(req.prefix or "", req.suffix or "")
         logger.info("[%s] llm_input_prefix=%r", request_id, llm_prefix)
         logger.info("[%s] llm_input_suffix=%r", request_id, llm_suffix)
-        prompt = build_prompt(req.prefix, req.suffix, req.languageId, location=location)
+        
+        prompt = build_prompt(
+            req.prefix, 
+            req.suffix, 
+            req.languageId, 
+            location=location,
+            thinking_level=req.model_thinking,
+            preferences=req.user_preferences
+        )
         result = await call_ollama(prompt, tag=f"{request_id}-primary", temperature=0.7)
 
         content = result["content"] or ""

backend/prompt.py

@@ -29,20 +29,46 @@ def prepare_prompt_context(prefix: str, suffix: str) -> Tuple[str, str]:
     return _prepare_context(prefix, suffix)
 
 
-def build_prompt(prefix: str, suffix: str, language_id: str = "markdown", location: str = "") -> str:
+def build_prompt(
+    prefix: str, 
+    suffix: str, 
+    language_id: str = "markdown", 
+    location: str = "",
+    thinking_level: str = "low",
+    preferences: object = None
+) -> str:
     safe_language_id = _sanitize_language_id(language_id)
     recent_prefix, recent_suffix = _prepare_context(prefix, suffix)
     current_time = _get_current_datetime()
     location_info = f"\nUser location: {location}" if location else ""
+    
+    thinking_instruction = ""
+    if thinking_level == "medium":
+        thinking_instruction = "\n- Briefly analyze the context before suggesting."
+    elif thinking_level == "high":
+        thinking_instruction = "\n- Deeply analyze the context, structure, and intent before suggesting. Think step-by-step."
 
-    prompt = f"""Current time: {current_time}{location_info}
+    pref_info = []
+    if preferences:
+        if preferences.language and preferences.language != 'auto':
+            pref_info.append(f"Preferred language: {preferences.language}")
+        if preferences.currency and preferences.currency != 'auto':
+            pref_info.append(f"Preferred currency: {preferences.currency}")
+        if preferences.timezone and preferences.timezone != 'auto':
+            pref_info.append(f"User timezone: {preferences.timezone}")
+    
+    preferences_instruction = "\n".join(pref_info)
+    if preferences_instruction:
+        preferences_instruction = f"\nUser Preferences:\n{preferences_instruction}"
+
+    prompt = f"""Current time: {current_time}{location_info}{preferences_instruction}
 
 You are an inline completion engine for a {safe_language_id} editor with ghost-text suggestions.
 
 Your job:
 - Return ONLY the text that should be inserted at the cursor between PREFIX and SUFFIX.
 - Prefer a meaningful, non-empty insertion with moderate length.
-- Avoid overly short outputs with little information value.
+- Avoid overly short outputs with little information value.{thinking_instruction}
 
 Important context:
 - PREFIX may contain OCR metadata inline after images, e.g. ![alt](url) <OCR:description>.