feat: Add Zernike GNN surrogate module and M1 mirror V12/V13 studies

This commit introduces the GNN-based surrogate for Zernike mirror optimization
and the M1 mirror study progression from V12 (GNN validation) to V13 (pure NSGA-II).

## GNN Surrogate Module (optimization_engine/gnn/)

New module for Graph Neural Network surrogate prediction of mirror deformations:

- `polar_graph.py`: PolarMirrorGraph - fixed 3000-node polar grid structure
- `zernike_gnn.py`: ZernikeGNN with design-conditioned message passing
- `differentiable_zernike.py`: GPU-accelerated Zernike fitting and objectives
- `train_zernike_gnn.py`: ZernikeGNNTrainer with multi-task loss
- `gnn_optimizer.py`: ZernikeGNNOptimizer for turbo mode (~900k trials/hour)
- `extract_displacement_field.py`: OP2 to HDF5 field extraction
- `backfill_field_data.py`: Extract fields from existing FEA trials

Key innovation: Design-conditioned convolutions that modulate message passing
based on structural design parameters, enabling accurate field prediction.

## M1 Mirror Studies

### V12: GNN Field Prediction + FEA Validation
- Zernike GNN trained on V10/V11 FEA data (238 samples)
- Turbo mode: 5000 GNN predictions → top candidates → FEA validation
- Calibration workflow for GNN-to-FEA error correction
- Scripts: run_gnn_turbo.py, validate_gnn_best.py, compute_full_calibration.py

### V13: Pure NSGA-II FEA (Ground Truth)
- Seeds 217 FEA trials from V11+V12
- Pure multi-objective NSGA-II without any surrogate
- Establishes ground-truth Pareto front for GNN accuracy evaluation
- Narrowed blank_backface_angle range to [4.0, 5.0]

## Documentation Updates

- SYS_14: Added Zernike GNN section with architecture diagrams
- CLAUDE.md: Added GNN module reference and quick start
- V13 README: Study documentation with seeding strategy

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

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

optimization_engine/gnn/test_new_extraction.py

@@ -0,0 +1,35 @@
+"""Test the fixed extraction function directly on OP2."""
+import sys
+sys.path.insert(0, "C:/Users/Antoine/Atomizer")
+
+import numpy as np
+from pathlib import Path
+from optimization_engine.gnn.extract_displacement_field import extract_displacement_field
+
+# Test direct extraction from OP2
+op2_path = Path("C:/Users/Antoine/Atomizer/studies/m1_mirror_adaptive_V11/2_iterations/iter91/assy_m1_assyfem1_sim1-solution_1.op2")
+
+print(f"Testing extraction from: {op2_path.name}")
+print(f"Exists: {op2_path.exists()}")
+
+if op2_path.exists():
+    field_data = extract_displacement_field(op2_path, r_inner=100.0, r_outer=650.0)
+
+    print(f"\n=== EXTRACTION RESULT ===")
+    print(f"Total surface nodes: {len(field_data['node_ids'])}")
+
+    coords = field_data['node_coords']
+    r = np.sqrt(coords[:, 0]**2 + coords[:, 1]**2)
+    print(f"Radial range: [{r.min():.1f}, {r.max():.1f}] mm")
+    print(f"Z range: [{coords[:, 2].min():.1f}, {coords[:, 2].max():.1f}] mm")
+
+    print(f"\nSubcases: {list(field_data['z_displacement'].keys())}")
+    for sc, disp in field_data['z_displacement'].items():
+        nan_count = np.sum(np.isnan(disp))
+        if nan_count == 0:
+            print(f"  Subcase {sc}: Z-disp range [{disp.min():.6f}, {disp.max():.6f}] mm")
+        else:
+            valid = disp[~np.isnan(disp)]
+            print(f"  Subcase {sc}: {nan_count}/{len(disp)} NaN values, valid range: [{valid.min():.6f}, {valid.max():.6f}]")
+else:
+    print("OP2 file not found!")