feat: Add substudy system with live history tracking and workflow fixes

Major Features:
- Hierarchical substudy system (like NX Solutions/Subcases)
  * Shared model files across all substudies
  * Independent configuration per substudy
  * Continuation support from previous substudies
  * Real-time incremental history updates
- Live history tracking with optimization_history_incremental.json
- Complete bracket_displacement_maximizing study with substudy examples

Core Fixes:
- Fixed expression update workflow to pass design_vars through simulation_runner
  * Restored working NX journal expression update mechanism
  * OP2 timestamp verification instead of file deletion
  * Resolved issue where all trials returned identical objective values
- Fixed LLMOptimizationRunner to pass design variables to simulation runner
- Enhanced NXSolver with timestamp-based file regeneration verification

New Components:
- optimization_engine/llm_optimization_runner.py - LLM-driven optimization runner
- optimization_engine/optimization_setup_wizard.py - Phase 3.3 setup wizard
- studies/bracket_displacement_maximizing/ - Complete substudy example
  * run_substudy.py - Substudy runner with continuation
  * run_optimization.py - Standalone optimization runner
  * config/substudy_template.json - Template for new substudies
  * substudies/coarse_exploration/ - 20-trial coarse search
  * substudies/fine_tuning/ - 50-trial refinement (continuation example)
  * SUBSTUDIES_README.md - Complete substudy documentation

Technical Improvements:
- Incremental history saving after each trial (optimization_history_incremental.json)
- Expression update workflow: .prt update → NX journal receives values → geometry update → FEM update → solve
- Trial indexing fix in substudy result saving
- Updated README with substudy system documentation

Testing:
- Successfully ran 20-trial coarse_exploration substudy
- Verified different objective values across trials (workflow fix validated)
- Confirmed live history updates in real-time
- Tested shared model file usage across substudies

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

docs/SESSION_SUMMARY_PHASE_3.md

@@ -6,13 +6,13 @@
 
 ## Overview
 
-Phase 3 implements **automated research and code generation** for OP2 result extraction using pyNastran. The system can:
+Phase 3 implements **LLM-enhanced research and code generation** for OP2 result extraction using pyNastran. The system can:
 1. Research pyNastran documentation to find appropriate APIs
 2. Generate complete, executable Python extraction code
 3. Store learned patterns in a knowledge base
 4. Auto-generate extractors from Phase 2.7 LLM output
 
-This completes the **zero-manual-coding vision**: Users describe optimization goals in natural language → System generates all required code automatically.
+This enables **LLM-enhanced optimization workflows**: Users can describe goals in natural language and optionally have the system generate code automatically, or write custom extractors manually as needed.
 
 ## Objectives Achieved
 
@@ -287,7 +287,7 @@ def min_to_avg_ratio_hook(context):
     return {'min_to_avg_ratio': result, 'objective': result}
 ```
 
-**Result**: Complete optimization setup from natural language → Zero manual coding! 🚀
+**Result**: LLM-enhanced optimization setup from natural language with flexible automation! 🚀
 
 ## Testing
 
@@ -483,13 +483,13 @@ Phase 3 successfully implements **automated OP2 extraction code generation** usi
 - ✅ Knowledge base architecture
 - ✅ 3 core extraction patterns (displacement, stress, force)
 
-This completes the **zero-manual-coding pipeline**:
+This enables the **LLM-enhanced automation pipeline**:
 - Phase 2.7: LLM analyzes natural language → engineering features
-- Phase 2.8: Inline calculation code generation
-- Phase 2.9: Post-processing hook generation
-- **Phase 3: OP2 extraction code generation**
+- Phase 2.8: Inline calculation code generation (optional)
+- Phase 2.9: Post-processing hook generation (optional)
+- **Phase 3: OP2 extraction code generation (optional)**
 
-Users can now describe optimization goals in natural language and the system generates ALL required code automatically! 🎉
+Users can describe optimization goals in natural language and choose to leverage automated code generation, manual coding, or a hybrid approach! 🎉
 
 ## Related Documentation
 

docs/SESSION_SUMMARY_PHASE_3_1.md

@@ -6,19 +6,19 @@
 
 ## Overview
 
-Phase 3.1 completes the **zero-manual-coding automation pipeline** by integrating:
+Phase 3.1 completes the **LLM-enhanced automation pipeline** by integrating:
 - **Phase 2.7**: LLM workflow analysis
 - **Phase 3.0**: pyNastran research agent
 - **Phase 2.8**: Inline code generation
 - **Phase 2.9**: Post-processing hook generation
 
-The result: Users describe optimization goals in natural language → System automatically generates ALL required code from request to execution!
+The result: Users can describe optimization goals in natural language and choose to leverage automatic code generation, manual coding, or a hybrid approach!
 
 ## Objectives Achieved
 
-### ✅ Complete Automation Pipeline
+### ✅ LLM-Enhanced Automation Pipeline
 
-**From User Request to Execution - Zero Manual Coding:**
+**From User Request to Execution - Flexible LLM-Assisted Workflow:**
 
 ```
 User Natural Language Request
@@ -298,7 +298,7 @@ Trial N completed
 Objective value: 0.072357
 ```
 
-**ZERO manual coding from user request to Optuna trial!** 🚀
+**LLM-enhanced workflow with optional automation from user request to Optuna trial!** 🚀
 
 ## Key Integration Points
 
@@ -429,9 +429,9 @@ Result: PASSED!
 
 ## Benefits
 
-### 1. Complete Automation
+### 1. LLM-Enhanced Flexibility
 
-**Before** (Manual workflow):
+**Traditional Manual Workflow**:
 ```
 1. User describes optimization
 2. Engineer manually writes OP2 extractor
@@ -441,32 +441,33 @@ Result: PASSED!
 Time: Hours to days
 ```
 
-**After** (Automated workflow):
+**LLM-Enhanced Workflow**:
 ```
 1. User describes optimization in natural language
-2. System generates ALL code automatically
-Time: Seconds
+2. System offers to generate code automatically OR user writes custom code
+3. Hybrid approach: mix automated and manual components as needed
+Time: Seconds to minutes (user choice)
 ```
 
-### 2. Zero Learning Curve
+### 2. Reduced Learning Curve
 
-Users don't need to know:
-- ❌ pyNastran API
-- ❌ OP2 file structure
-- ❌ Python coding
-- ❌ Optimization framework
+LLM assistance helps users who are unfamiliar with:
+- pyNastran API (can still write custom extractors if desired)
+- OP2 file structure (LLM provides templates)
+- Python coding best practices (LLM generates examples)
+- Optimization framework patterns (LLM suggests approaches)
 
-They only need to describe **what they want** in natural language!
+Users can describe goals in natural language and choose their preferred level of automation!
 
-### 3. Correct by Construction
+### 3. Quality LLM-Generated Code
 
-Generated code uses:
+When using automated generation, code uses:
 - ✅ Proven extraction patterns from research agent
 - ✅ Correct API paths from documentation
 - ✅ Proper data structure access
 - ✅ Error handling and validation
 
-No manual bugs!
+Users can review, modify, or replace generated code as needed!
 
 ### 4. Extensible
 
@@ -570,10 +571,10 @@ None - Phase 3.1 is purely additive!
 
 ## Summary
 
-Phase 3.1 successfully completes the **zero-manual-coding automation pipeline**:
+Phase 3.1 successfully completes the **LLM-enhanced automation pipeline**:
 
 - ✅ Orchestrator integrates Phase 2.7 + Phase 3.0
-- ✅ Automatic extractor generation from LLM output
+- ✅ Optional automatic extractor generation from LLM output
 - ✅ Dynamic loading and execution on real OP2 files
 - ✅ Smart parameter filtering per pattern type
 - ✅ Multi-extractor support
@@ -581,28 +582,28 @@ Phase 3.1 successfully completes the **zero-manual-coding automation pipeline**:
 - ✅ Extraction successful: max_disp=0.361783mm
 - ✅ Normalized objective calculated: 0.072357
 
-**Complete Automation Verified:**
+**LLM-Enhanced Workflow Verified:**
 ```
 Natural Language Request
     ↓
 Phase 2.7 LLM → Engineering Features
     ↓
-Phase 3.1 Orchestrator → Generated Extractors
+Phase 3.1 Orchestrator → Generated Extractors (or manual extractors)
     ↓
-Phase 3.0 Research Agent → OP2 Extraction Code
+Phase 3.0 Research Agent → OP2 Extraction Code (optional)
     ↓
 Execution on Real OP2 → Results
     ↓
-Phase 2.8 Inline Calc → Calculations
+Phase 2.8 Inline Calc → Calculations (optional)
     ↓
-Phase 2.9 Hooks → Objective Value
+Phase 2.9 Hooks → Objective Value (optional)
     ↓
 Optuna Trial Complete
 
-ZERO MANUAL CODING! 🚀
+LLM-ENHANCED WITH USER FLEXIBILITY! 🚀
 ```
 
-Users can now describe optimization goals in natural language and the system automatically generates and executes ALL required code from request to final objective value!
+Users can describe optimization goals in natural language and choose to leverage automated code generation, write custom code, or use a hybrid approach as needed!
 
 ## Related Documentation