Add API usage tracking and dynamic task reloading

Features:
- Usage tracking system (usage_tracker.py)
  - Tracks input/output tokens per API call
  - Calculates costs with support for cache pricing
  - Stores data in usage_data.json (gitignored)
  - Integrated into llm_interface.py

- Dynamic task scheduler reloading
  - Auto-detects YAML changes every 60s
  - No restart needed for new tasks
  - reload_tasks() method for manual refresh

- Example cost tracking scheduled task
  - Daily API usage report
  - Budget tracking ($5/month target)
  - Disabled by default in scheduled_tasks.yaml

Improvements:
- Fixed tool_use/tool_result pair splitting bug (CRITICAL)
- Added thread safety to agent.chat()
- Fixed N+1 query problem in hybrid search
- Optimized database batch queries
- Added conversation history pruning (50 messages max)

Updated .gitignore:
- Exclude user profiles (memory_workspace/users/*.md)
- Exclude usage data (usage_data.json)
- Exclude vector index (vectors.usearch)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

llm_interface.py

@@ -7,6 +7,8 @@ import requests
 from anthropic import Anthropic
 from anthropic.types import Message
 
+from usage_tracker import UsageTracker
+
 # API key environment variable names by provider
 _API_KEY_ENV_VARS = {
     "claude": "ANTHROPIC_API_KEY",
@@ -29,6 +31,7 @@ class LLMInterface:
         self,
         provider: str = "claude",
         api_key: Optional[str] = None,
+        track_usage: bool = True,
     ) -> None:
         self.provider = provider
         self.api_key = api_key or os.getenv(
@@ -37,6 +40,9 @@ class LLMInterface:
         self.model = _DEFAULT_MODELS.get(provider, "")
         self.client: Optional[Anthropic] = None
 
+        # Usage tracking
+        self.tracker = UsageTracker() if track_usage else None
+
         if provider == "claude":
             self.client = Anthropic(api_key=self.api_key)
 
@@ -46,7 +52,11 @@ class LLMInterface:
         system: Optional[str] = None,
         max_tokens: int = 4096,
     ) -> str:
-        """Send chat request and get response."""
+        """Send chat request and get response.
+
+        Raises:
+            Exception: If the API call fails or returns an unexpected response.
+        """
         if self.provider == "claude":
             response = self.client.messages.create(
                 model=self.model,
@@ -54,6 +64,23 @@ class LLMInterface:
                 system=system or "",
                 messages=messages,
             )
+
+            # Track usage
+            if self.tracker and hasattr(response, "usage"):
+                self.tracker.track(
+                    model=self.model,
+                    input_tokens=response.usage.input_tokens,
+                    output_tokens=response.usage.output_tokens,
+                    cache_creation_tokens=getattr(
+                        response.usage, "cache_creation_input_tokens", 0
+                    ),
+                    cache_read_tokens=getattr(
+                        response.usage, "cache_read_input_tokens", 0
+                    ),
+                )
+
+            if not response.content:
+                return ""
             return response.content[0].text
 
         if self.provider == "glm":
@@ -67,7 +94,9 @@ class LLMInterface:
             headers = {"Authorization": f"Bearer {self.api_key}"}
             response = requests.post(
                 _GLM_BASE_URL, json=payload, headers=headers,
+                timeout=60,
             )
+            response.raise_for_status()
             return response.json()["choices"][0]["message"]["content"]
 
         raise ValueError(f"Unsupported provider: {self.provider}")
@@ -111,8 +140,37 @@ class LLMInterface:
             messages=messages,
             tools=tools,
         )
+
+        # Track usage
+        if self.tracker and hasattr(response, "usage"):
+            self.tracker.track(
+                model=self.model,
+                input_tokens=response.usage.input_tokens,
+                output_tokens=response.usage.output_tokens,
+                cache_creation_tokens=getattr(
+                    response.usage, "cache_creation_input_tokens", 0
+                ),
+                cache_read_tokens=getattr(
+                    response.usage, "cache_read_input_tokens", 0
+                ),
+            )
+
         return response
 
     def set_model(self, model: str) -> None:
         """Change the active model."""
         self.model = model
+
+    def get_usage_stats(self, target_date: Optional[str] = None) -> Dict:
+        """Get usage statistics and costs.
+
+        Args:
+            target_date: Date string (YYYY-MM-DD). If None, returns today's stats.
+
+        Returns:
+            Dict with cost, token counts, and breakdown by model.
+        """
+        if not self.tracker:
+            return {"error": "Usage tracking not enabled"}
+
+        return self.tracker.get_daily_cost(target_date)