feat: Add learning cycle and validation tracking

Priority 2 & 3 Implementation:
- Learning cycle tracking: Wrap run_learning_cycle() with track_task_execution()
- Validation tracking: Wrap ValidationEngine.run() with track_task_execution()
- Both use silent tracking (communicate_estimate=False) for internal data collection
- Builds historical data for future time estimation improvements

This completes the minimal viable self-tracking implementation:
1. Chat response integration (user-facing)
2. Learning cycle tracking (build data)
3. Validation tracking (build data)

Author: anhmtk

stillme_core/learning/scheduler.py

@@ -119,12 +119,28 @@ async def run_learning_cycle(self) -> Dict[str, Any]:
         Returns:
             Dict with cycle results: cycle_number, entries_fetched, entries_added_to_rag, etc.
         """
-        start_time = datetime.now()
         cycle_number = self.cycle_count + 1
 
-        logger.info(f"🔄 Starting learning cycle #{cycle_number}...")
+        # Track learning cycle execution time
+        from stillme_core.monitoring import track_task_execution
 
-        try:
+        # Determine complexity and size (will be updated after fetch)
+        complexity = "moderate"
+        size = 50  # Default estimate
+        
+        with track_task_execution(
+            task_description=f"Learning cycle #{cycle_number}",
+            task_type="learning",
+            complexity=complexity,
+            size=size,
+            metadata={"cycle_number": cycle_number},
+            communicate_estimate=False  # Internal tracking only
+        ):
+            start_time = datetime.now()
+            
+            logger.info(f"🔄 Starting learning cycle #{cycle_number}...")
+            
+            try:
             # Step 1: Fetch from all sources
             all_entries = self.source_integration.fetch_all_sources(
                 max_items_per_source=5,
@@ -234,11 +250,11 @@ async def run_learning_cycle(self) -> Dict[str, Any]:
                 "status": "success"
             }
 
-            logger.info(f"✅ Learning cycle #{cycle_number} completed: {entries_added_to_rag} entries added to RAG in {processing_time:.2f}s")
-            
-            return result
-            
-        except Exception as e:
+                logger.info(f"✅ Learning cycle #{cycle_number} completed: {entries_added_to_rag} entries added to RAG in {processing_time:.2f}s")
+                
+                return result
+                
+            except Exception as e:
             logger.error(f"❌ Error in learning cycle #{cycle_number}: {e}", exc_info=True)
 
             # Record error in unified metrics

stillme_core/validation/chain.py

@@ -107,11 +107,29 @@ def run(self, answer: str, ctx_docs: List[str], context_quality: Optional[str] =
         Returns:
             ValidationResult with overall status
         """
-        reasons: List[str] = []
-        patched = answer
-        has_citation = False
-        low_overlap_only = False
-        has_critical_failure = False  # Track if any critical validator failed without patch
+        # Track validation execution time (internal tracking)
+        from stillme_core.monitoring import track_task_execution
+        
+        num_validators = len(self.validators)
+        response_length = len(answer)
+        
+        with track_task_execution(
+            task_description=f"Validation: {num_validators} validators",
+            task_type="validation",
+            complexity="moderate",
+            size=response_length,
+            metadata={
+                "num_validators": num_validators,
+                "num_context_docs": len(ctx_docs),
+                "is_philosophical": is_philosophical
+            },
+            communicate_estimate=False  # Internal tracking only
+        ):
+            reasons: List[str] = []
+            patched = answer
+            has_citation = False
+            low_overlap_only = False
+            has_critical_failure = False  # Track if any critical validator failed without patch
 
         # Group validators into sequential and parallel groups
         sequential_validators = []