z-score warning

[jgallowa07]

I don't know if I would consider this a "bug", but whenever I run zscore I get

/home/jared/MatsenGroup/Projects/phippery/phippery/phippery/zscore.py:125: RuntimeWarning: invalid value encountered in divide
  zs = (data - null_means) / null_stds
/home/jared/MatsenGroup/Projects/phippery/phippery/phippery/zscore.py:125: RuntimeWarning: divide by zero encountered in divide
  zs = (data - null_means) / null_stds

I've tried doing a safer divide strategy by replacing this chunk with:

        else:
            data = data_df.loc[pid].to_numpy()
            null_means = means_df.loc[ibin].to_numpy()
            null_stds = stds_df.loc[ibin].to_numpy()
            # Replace invalid values in null_stds with NaN
            null_stds = np.where(
                np.isfinite(null_stds) & (null_stds != 0), null_stds, np.nan
            )

            # Compute zs safely; np.nan_to_num replaces NaNs with 0 in the result
            zs = np.nan_to_num((data - null_means) / null_stds, nan=0.0)

        assert zs.dtype == np.float64
        zscore_df.loc[pid] = zs

    return zscore_df, means_df, stds_df

but for some strange reason the enrichment table comes out as an object datatype in the xarray object. This isn't really problem so instead I'll just be adding a warning catch to #189