numcodecs codecs do not get dictified in metadata.to_dict()

[d-v-b]

# /// script
# requires-python = ">=3.11"
# dependencies = [
#     "zarr",
# ]
# ///


import zarr
import numcodecs

array = zarr.create_array(
    {}, 
    shape=(10,), 
    zarr_format=2, 
    compressors = (numcodecs.LZMA()),
    dtype='uint8')
print(array.metadata.to_dict()["compressor"])
# LZMA(format=1, check=-1, preset=None, filters=None)
# should be {"id": "lzma", ...}

IMO the best fix is to rethink how we handle in-house vs numcodecs codecs.