🚫 Fix trace upload failure by removing evaluation module dependency

✅ Problem solved:
• Fixed 'No module named evaluation' error in trace upload
• Removed problematic imports causing circular dependency chain

�� Root cause:
• agentgraph/__init__.py imported SlidingWindowMonitor
• This triggered import chain: extraction → methods → baseline → evaluation
• evaluation module doesn't exist in HF Spaces deployment

⚡ Fix applied:
• Commented out SlidingWindowMonitor and related imports in agentgraph/__init__.py
• Removed problematic __all__ exports
• Kept essential trace_management functionality intact

🧪 Local testing: ✅ All passed
• trace_management import: ✅
• analyze_trace_characteristics: ✅ (structured_data type detected)
• Database save_trace: ✅ (ID: 1, trace_id: 8bdedf4b...)

🚀 Result: Trace file upload should work in HF Spaces now
📋 Impact: Only affects advanced features, core trace upload remains functional

agentgraph/__init__.py

@@ -33,14 +33,15 @@ from .input import (
     display_trace_summary,
     preprocess_content_for_cost_optimization
 )
-from .extraction import SlidingWindowMonitor
-from .reconstruction import (
-    PromptReconstructor, 
-    reconstruct_prompts_from_knowledge_graph,
-    enrich_knowledge_graph_with_prompts as enrich_reconstruction_graph
-)
-from .testing import run_knowledge_graph_tests
-from .causal import analyze_causal_effects, enrich_knowledge_graph as enrich_causal_graph, generate_report as generate_causal_report
+# NOTE: SlidingWindowMonitor removed to avoid importing evaluation module
+# from .extraction import SlidingWindowMonitor
+# from .reconstruction import (
+#     PromptReconstructor, 
+#     reconstruct_prompts_from_knowledge_graph,
+#     enrich_knowledge_graph_with_prompts as enrich_reconstruction_graph
+# )
+# from .testing import run_knowledge_graph_tests
+# from .causal import analyze_causal_effects, enrich_knowledge_graph as enrich_causal_graph, generate_report as generate_causal_report
 
 # Import parser system for platform-specific trace analysis
 from .input.parsers import (
@@ -57,21 +58,21 @@ __version__ = "0.1.0"
 __all__ = [
     # Core components
     'ChunkingService',
-    'SlidingWindowMonitor',
-    'PromptReconstructor',
-    'run_knowledge_graph_tests',
-    'analyze_causal_effects',
-    'enrich_causal_graph',
-    'generate_causal_report',
+    # 'SlidingWindowMonitor',  # Removed to avoid evaluation module import
+    # 'PromptReconstructor',
+    # 'run_knowledge_graph_tests',
+    # 'analyze_causal_effects',
+    # 'enrich_causal_graph',
+    # 'generate_causal_report',
     
     # Input analysis functions
     'analyze_trace_characteristics',
     'display_trace_summary',
     'preprocess_content_for_cost_optimization',
     
-    # Reconstruction functions
-    'reconstruct_prompts_from_knowledge_graph',
-    'enrich_reconstruction_graph',
+    # Reconstruction functions - commented out to avoid imports
+    # 'reconstruct_prompts_from_knowledge_graph',
+    # 'enrich_reconstruction_graph',
     
     # Parser system
     'BaseTraceParser', 'LangSmithParser', 'ParsedMetadata',
@@ -80,5 +81,5 @@ __all__ = [
     
     # Shared models and utilities
     'Entity', 'Relation', 'KnowledgeGraph',
-    'ContentReference', 'Failure', 'Report'
+    'ContentReference', 'Failure', # 'Report' - might not exist
 ]