Can't read empty categorical column inside a parquet file

[Vincenthays]

Polars version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of Polars.

Issue description

The default engine crash when reading an empty categorical column, whereas pyarrow is working as expected

Reproducible example

import polars as pl

pl.DataFrame({'a': pl.Series([None], dtype=pl.Categorical)}).write_parquet('test.pq')

pl.read_parquet('test.pq', use_pyarrow=True) # working with pyarrow
pl.read_parquet('test.pq')                   # not working without pyarrow
pl.scan_parquet('test.pq').collect()         # not working without pyarrow

Expected behavior

>> import polars as pl
>> pl.DataFrame({'a': pl.Series([None], dtype=pl.Categorical)}).write_parquet('test.pq')
>> pl.read_parquet('test.pq')
shape: (1, 1)
│   a   │
│ ----- │
│  cat  │
│═══════│ 
|  null |
>> pl.scan_parquet('test.pq').collect()
shape: (1, 1)
│   a   │
│ ----- │
│  cat  │
│═══════│ 
|  null |

Installed versions

---Version info---
Polars: 0.15.1
Index type: UInt32
Platform: macOS-13.0.1-arm64-arm-64bit
Python: 3.10.8 (main, Oct 13 2022, 09:48:40) [Clang 14.0.0 (clang-1400.0.29.102)]
---Optional dependencies---
pyarrow: 10.0.1
pandas: 1.5.2
numpy: 1.22.4
fsspec: 2022.11.0
connectorx: 0.3.1
xlsx2csv: <not installed>
matplotlib: <not installed>