Handle None values when writing and avoid AnnData file corruption.

[SeppeDeWinter]

Hi

Thank you for the impressive work!

When a None value is trying to be written to the AnnData file, the program will crash resulting in a corrupted AnnData.

For example:

Polars allows for None values https://docs.pola.rs/user-guide/expressions/missing-data/ and when a Polars Series of type str is trying to be written to the AnnData file the program will panic resulting in a corrupted hdf5 file.

thread '<unnamed>' panicked at /root/.cargo/registry/src/index.crates.io-6f17d22bba15001f/anndata-0.4.1/src/data/array/dataframe.rs:261:28: called `Option::unwrap()` on a `None` value

Reproducible example

import polars as pl
import snapatac2 as snap

adata = snap.AnnData(
        filename = "/tmp/test.h5ad",
        obs = pl.DataFrame({"Index": ["foo1", "foo2", "foo3", "foo4", "foo5"]})
)

adata.obs["test"] = pl.Series(name = "example", values = ["bar1", "bar2", "bar3", "bar4", None])

thread '<unnamed>' panicked at /root/.cargo/registry/src/index.crates.io-6f17d22bba15001f/anndata-0.4.1/src/data/array/dataframe.rs:261:28:
called `Option::unwrap()` on a `None` value
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Aborted (core dumped)

File corruption

>>> import snapatac2 as snap
>>> snap.read("/tmp/test.h5ad")
thread '<unnamed>' panicked at /root/.cargo/registry/src/index.crates.io-6f17d22bba15001f/pyanndata-0.4.1/src/anndata.rs:36:60:
called `Result::unwrap()` on an `Err` value: H5Fopen(): unable to open file: bad object header version number
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
---------------------------------------------------------------------------
PanicException                            Traceback (most recent call last)
Cell In[2], line 1
----> 1 snap.read("/tmp/test.h5ad")

PanicException: called `Result::unwrap()` on an `Err` value: H5Fopen(): unable to open file: bad object header version number

This is caused by the unhandled exceptions (unwrap) at:

https://github.com/kaizhang/anndata-rs/blob/a7752429597126c4eb548ebb064dda717294c7e8/anndata/src/data/array/dataframe.rs#L175-L275

All the best,

Seppe