feat(scenegraph): add depth-based 3D positioning via pinhole projection

- Add bbox_depth_center position strategy in SceneGraphBuilder using depth
  at bbox centre and configurable camera_hfov_degrees for pinhole projection.
- Add optional depth_sensor_uuid to HabitatSimulatorConfig; create depth
  sensor spec alongside RGB sensor.
- Add camera_position/camera_rotation fields to RoomView; capture pose from
  sensor_states when depth sensor is available.
- Update flatten_room_views for backward compatibility with legacy tuple
  format.
- Wired in depth sensor and bbox_depth_center strategy in verification
  notebook.
- Add tests for depth sensor support and new position strategies.

mini-nav/simulator/habitat.py

@@ -15,6 +15,26 @@ class HabitatSimulatorConfig:
     enable_physics: bool = False
     sensor_uuid: str = "color_sensor"
     agent_id: int = 0
+    depth_sensor_uuid: str | None = "depth_sensor"
+    hfov_degrees: float = 90.0
+
+
+def _camera_sensor_spec(
+    habitat_sim_module: Any,
+    *,
+    uuid: str,
+    sensor_type: Any,
+    image_size: int,
+    sensor_height: float,
+    hfov_degrees: float,
+) -> Any:
+    spec = habitat_sim_module.CameraSensorSpec()
+    spec.uuid = uuid
+    spec.sensor_type = sensor_type
+    spec.resolution = [image_size, image_size]
+    spec.position = [0.0, sensor_height, 0.0]
+    spec.hfov = float(hfov_degrees)
+    return spec
 
 
 def create_habitat_simulator(
@@ -30,6 +50,11 @@ def create_habitat_simulator(
     if config.move_forward_step <= 0:
         raise ValueError("move_forward_step must be greater than 0")
 
+    if not (0 < config.hfov_degrees < 180):
+        raise ValueError(
+            f"hfov_degrees must be in (0, 180), got {config.hfov_degrees}"
+        )
+
     if habitat_sim_module is None:
         habitat_sim_module = import_module("habitat_sim")
 
@@ -38,12 +63,31 @@ def create_habitat_simulator(
     sim_cfg.enable_physics = config.enable_physics
 
     agent_cfg = habitat_sim_module.agent.AgentConfiguration()
-    rgb_sensor_spec = habitat_sim_module.CameraSensorSpec()
-    rgb_sensor_spec.uuid = config.sensor_uuid
-    rgb_sensor_spec.sensor_type = habitat_sim_module.SensorType.COLOR
-    rgb_sensor_spec.resolution = [config.image_size, config.image_size]
-    rgb_sensor_spec.position = [0.0, config.sensor_height, 0.0]
-    agent_cfg.sensor_specifications = [rgb_sensor_spec]
+
+    sensor_specs = [
+        _camera_sensor_spec(
+            habitat_sim_module,
+            uuid=config.sensor_uuid,
+            sensor_type=habitat_sim_module.SensorType.COLOR,
+            image_size=config.image_size,
+            sensor_height=config.sensor_height,
+            hfov_degrees=config.hfov_degrees,
+        ),
+    ]
+
+    if config.depth_sensor_uuid is not None:
+        sensor_specs.append(
+            _camera_sensor_spec(
+                habitat_sim_module,
+                uuid=config.depth_sensor_uuid,
+                sensor_type=habitat_sim_module.SensorType.DEPTH,
+                image_size=config.image_size,
+                sensor_height=config.sensor_height,
+                hfov_degrees=config.hfov_degrees,
+            ),
+        )
+
+    agent_cfg.sensor_specifications = sensor_specs
 
     turn_angle = 360.0 / config.views_per_room
     agent_cfg.action_space = {