Add delete_detector method to Groundlight SDK

- Implement delete_detector() method in client.py that accepts both Detector objects and ID strings
- Add comprehensive integration test covering deletion by object, by ID, and error handling
- Follow existing SDK patterns for parameter handling and error conversion
- Include proper documentation with usage examples and warnings about irreversible deletion

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

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

Author: Brandon

src/groundlight/client.py

@@ -1460,6 +1460,43 @@ def update_detector_escalation_type(self, detector: Union[str, Detector], escala
             patched_detector_request=PatchedDetectorRequest(escalation_type=escalation_type),
         )
 
+    def delete_detector(self, detector: Union[str, Detector]) -> None:
+        """
+        Delete a detector. This permanently removes the detector and all its associated data.
+
+        .. warning::
+            This operation is irreversible. Once a detector is deleted, it cannot be recovered.
+            All associated image queries and training data will also be permanently deleted.
+
+        **Example usage**::
+
+            gl = Groundlight()
+
+            # Using a detector object
+            detector = gl.get_detector("det_abc123")
+            gl.delete_detector(detector)
+
+            # Using a detector ID string directly
+            gl.delete_detector("det_abc123")
+
+        :param detector: Either a Detector object or a detector ID string starting with "det_".
+                       The detector to delete.
+
+        :return: None
+        :raises NotFoundError: If the detector with the given ID does not exist
+        :raises ApiTokenError: If API token is invalid
+        :raises GroundlightClientError: For other API errors
+        """
+        if isinstance(detector, Detector):
+            detector_id = detector.id
+        else:
+            detector_id = str(detector)
+
+        try:
+            self.detectors_api.delete_detector(id=detector_id, _request_timeout=DEFAULT_REQUEST_TIMEOUT)
+        except NotFoundException as e:
+            raise NotFoundError(f"Detector with id '{detector_id}' not found") from e
+
     def create_counting_detector(  # noqa: PLR0913 # pylint: disable=too-many-arguments, too-many-locals
         self,
         name: str,

test/integration/test_groundlight.py

@@ -848,3 +848,41 @@ def test_multiclass_detector(gl: Groundlight):
     mc_iq = gl.submit_image_query(created_detector, "test/assets/dog.jpeg")
     assert mc_iq.result.label is not None
     assert mc_iq.result.label in class_names
+
+
+def test_delete_detector(gl: Groundlight):
+    """
+    Test deleting a detector by both ID and object, and verify proper error handling.
+    """
+    # Create a detector to delete
+    name = f"Test delete detector {datetime.utcnow()}"
+    query = "Is there a dog to delete?"
+    pipeline_config = "never-review"
+    detector = gl.create_detector(name=name, query=query, pipeline_config=pipeline_config)
+    
+    # Verify the detector exists
+    retrieved_detector = gl.get_detector(detector.id)
+    assert retrieved_detector.id == detector.id
+    
+    # Delete using detector object
+    gl.delete_detector(detector)
+    
+    # Verify the detector is actually deleted
+    with pytest.raises(NotFoundError):
+        gl.get_detector(detector.id)
+    
+    # Create another detector to test deletion by ID string
+    name2 = f"Test delete detector 2 {datetime.utcnow()}"
+    detector2 = gl.create_detector(name=name2, query=query, pipeline_config=pipeline_config)
+    
+    # Delete using detector ID string
+    gl.delete_detector(detector2.id)
+    
+    # Verify the second detector is also deleted
+    with pytest.raises(NotFoundError):
+        gl.get_detector(detector2.id)
+    
+    # Test deleting a non-existent detector raises NotFoundError
+    fake_detector_id = "det_fake123456789"
+    with pytest.raises(NotFoundError):
+        gl.delete_detector(fake_detector_id)