refactor: Archive experimental LLM features for MVP stability (Phase 1.1)

Moved experimental LLM integration code to optimization_engine/future/:
- llm_optimization_runner.py - Runtime LLM API runner
- llm_workflow_analyzer.py - Workflow analysis
- inline_code_generator.py - Auto-generate calculations
- hook_generator.py - Auto-generate hooks
- report_generator.py - LLM report generation
- extractor_orchestrator.py - Extractor orchestration

Added comprehensive optimization_engine/future/README.md explaining:
- MVP LLM strategy (Claude Code skills, not runtime LLM)
- Why files were archived
- When to revisit post-MVP
- Production architecture reference

Production runner confirmed: optimization_engine/runner.py is sole active runner.

This establishes clear separation between:
- Production code (stable, no runtime LLM dependencies)
- Experimental code (archived for post-MVP exploration)

Part of Phase 1: Core Stabilization & Organization for MVP

Generated with Claude Code

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

optimization_engine/realtime_tracking.py

@@ -98,13 +98,25 @@ class RealtimeTrackingCallback:
 
     def _write_optimizer_state(self, study: optuna.Study, trial: optuna.trial.FrozenTrial):
         """Write current optimizer state."""
+        # [Protocol 11] For multi-objective, strategy is always NSGA-II
+        is_multi_objective = len(study.directions) > 1
+
+        if is_multi_objective:
+            # Multi-objective studies use NSGA-II, skip adaptive characterization
+            current_strategy = "NSGA-II"
+            current_phase = "multi_objective_optimization"
+        else:
+            # Single-objective uses intelligent strategy selection
+            current_strategy = getattr(self.optimizer, 'current_strategy', 'unknown')
+            current_phase = getattr(self.optimizer, 'current_phase', 'unknown')
+
         state = {
             "timestamp": datetime.now().isoformat(),
             "trial_number": trial.number,
             "total_trials": len(study.trials),
-            "current_phase": getattr(self.optimizer, 'current_phase', 'unknown'),
-            "current_strategy": getattr(self.optimizer, 'current_strategy', 'unknown'),
-            "is_multi_objective": len(study.directions) > 1,
+            "current_phase": current_phase,
+            "current_strategy": current_strategy,
+            "is_multi_objective": is_multi_objective,
             "study_directions": [str(d) for d in study.directions],
         }
 
@@ -132,18 +144,27 @@ class RealtimeTrackingCallback:
         else:
             log = []
 
+        # [Protocol 11] Handle both single and multi-objective
+        is_multi_objective = len(study.directions) > 1
+
         # Append new trial
         trial_entry = {
             "trial_number": trial.number,
             "timestamp": datetime.now().isoformat(),
             "state": str(trial.state),
             "params": trial.params,
-            "value": trial.value if trial.value is not None else None,
-            "values": trial.values if hasattr(trial, 'values') and trial.values is not None else None,
             "duration_seconds": (trial.datetime_complete - trial.datetime_start).total_seconds() if trial.datetime_complete else None,
             "user_attrs": dict(trial.user_attrs) if trial.user_attrs else {}
         }
 
+        # Add objectives (Protocol 11 compliant)
+        if is_multi_objective:
+            trial_entry["values"] = trial.values if trial.values is not None else None
+            trial_entry["value"] = None  # Not available
+        else:
+            trial_entry["value"] = trial.value if trial.value is not None else None
+            trial_entry["values"] = None
+
         log.append(trial_entry)
         self._atomic_write(trial_log_file, log)