Field data and geometry ======================= Cuts sample a field along one varying coordinate, while grids sample a field along two coordinates. Both store complex field coefficients in NumPy arrays. Geometry describes the coordinate surface; it is separate from the choice of field components. Some TICRA cut codes have more than one interpretation. An explicit ``CutGeometry`` tells the parser whether a cut is spherical, planar, on a surface, or cylindrical. Without that context, a cut's ``geometry`` and ``type`` may be ``None``. Grid parsing similarly accepts an optional ``GridGeometry``. The coefficient arrays use sample coordinates as their leading axes and the field component as their final axis. Grid arrays have shape ``(num_of_values_along_y, num_of_values_along_x, component_count)``. Missing grid samples can be represented by NumPy ``nan`` values.