Error when model is not compatible

[stakach]

i.e. trying to run a tensor model on a CPU

error callback information

2023-12-23T11:22:46.406832Z  ERROR - tensorflow_lite: Encountered unresolved custom op: %s.

then frame processing errors

2023-12-23T11:23:01.725975Z   WARN - tensorflow_lite: processing frame
no tensor data allocated (Exception)
  from /home/steve/projects/edge-ai/lib/tensorflow_lite/src/tensorflow_lite/tensor.cr:144:5 in 'as_u8'
  from /home/steve/projects/edge-ai/lib/tflite_image/src/tflite_image.cr:173:18 in 'apply_canvas_to_input_tensor'
  from /home/steve/projects/edge-ai/lib/tflite_image/src/tflite_image/object_detection.cr:13:5 in 'process'
  from /home/steve/projects/edge-ai/lib/tflite_pipeline/src/tflite_pipeline/coordinator.cr:130:17 in '->'
  from /home/steve/projects/edge-ai/lib/promise/src/promise/implicit_defer.cr:13:32 in '->'
  from /usr/share/crystal/src/fiber.cr:146:11 in 'run'
  from /usr/share/crystal/src/fiber.cr:98:34 in '->'
  from ???

We should not start the processing and expose information on why this is not running.