feat(cli): add depth verify/refine outputs and tests

- Retrieve depth + confidence measures from SVOReader when depth enabled
- Compute depth residual metrics and attach to output JSON
- Optionally write per-corner residual CSV via --report-csv
- Post-process refinement: optimize final pose and report pre/post metrics
- Add unit tests for depth verification and refinement modules
- Add basedpyright dev dependency for diagnostics

py_workspace/aruco/depth_refine.py

@@ -1,5 +1,5 @@
 import numpy as np
-from typing import Dict, Tuple, Optional
+from typing import Dict, Tuple
 from scipy.optimize import minimize
 from .pose_math import rvec_tvec_to_matrix, matrix_to_rvec_tvec
 from .depth_verify import compute_depth_residual
@@ -44,7 +44,7 @@ def depth_residual_objective(
     translation_diff = np.linalg.norm(param_diff[3:])
     regularization = regularization_weight * (rotation_diff + translation_diff)
 
-    return data_term + regularization
+    return float(np.real(data_term + regularization))
 
 
 def refine_extrinsics_with_depth(
@@ -55,7 +55,7 @@ def refine_extrinsics_with_depth(
     max_translation_m: float = 0.05,
     max_rotation_deg: float = 5.0,
     regularization_weight: float = 0.1,
-) -> Tuple[np.ndarray, dict]:
+) -> Tuple[np.ndarray, dict[str, float]]:
     initial_params = extrinsics_to_params(T_initial)
 
     max_rotation_rad = np.deg2rad(max_rotation_deg)