Migrate to pydantic v2

[wolny]

@sdvillal unfortunately the migration to v2 is not that simple. After upgrading geojson-pydantic installing pydantic-extra-types (needed for additional types like Color which we use in AnnotationModel and running the bump-pydantic migration tool, we're getting the same errors as before when creating a mock dataset:

Traceback (most recent call last):
  File "/Users/adrianwolny/workspace/pado/pado/demo.py", line 2, in <module>
    ds = mock_dataset(None)
  File "/Users/adrianwolny/workspace/pado/pado/mock.py", line 98, in mock_dataset
    ip = mock_images(images_urlpath, num_images)
  File "/Users/adrianwolny/workspace/pado/pado/mock.py", line 206, in mock_images
    image = Image(
  File "/Users/adrianwolny/workspace/pado/pado/images/image.py", line 149, in __init__
    self._metadata = self._load_metadata()
  File "/Users/adrianwolny/workspace/pado/pado/images/image.py", line 336, in _load_metadata
    return ImageMetadata(
  File "/Users/adrianwolny/miniconda3/envs/pasu/lib/python3.9/site-packages/pydantic/main.py", line 171, in __init__
    self.__pydantic_validator__.validate_python(data, self_instance=self)
pydantic_core._pydantic_core.ValidationError: 1 validation error for ImageMetadata
objective_power
  Input should be a valid string [type=string_type, input_value=40, input_type=int]
    For further information visit https://errors.pydantic.dev/2.6/v/string_type