fix counts column in points layer plus allocation

[lopollar]

Check if we can add columns to the points layer (Which I think is possible because of the gene column).

Then, if this column is added, adapt the allocation:
replace
cell_counts = combined_partitions.groupby(["cells", "gene"]).size()

by
if 'counts' in df.columns:
cell_counts=df.groupby(["cells", "gene"]).sum()['counts']
else:
cell_counts = df.groupby(["cells", "gene"]).size()

and that should be it!