Separating this out into its own issue. > Wondering if the dtype mismatch could be due to [this deactivation of some CF to Codecs logic](https://github.com/zarr-developers/VirtualiZarr/blob/ebed4006b2db8f0b48da0a2d4310239292400654/virtualizarr/parsers/hdf/hdf.py#L60C1-L69C28). _Originally posted by @jbusecke in [#196](https://github.com/NASA-IMPACT/veda-odd/issues/196#issuecomment-3028676162)_