Handle race conditions in apply_registration_to_image table copying

[jluethi]

At the end of the apply_registration_to_image task, we copy the tables from the reference cycle. This can lead to race conditions when the overwrite_input option is on, because every cycle copies tables from the reference cycle.

I thought I had this handled with:

            while current_round < max_retries:
                try:
                    old_table_group = zarr.open_group(
                        table_dict[table], mode="r"
                    )
                    current_round = max_retries
                except zarr.errors.GroupNotFoundError:
                    logger.debug(
                        f"Table {table} not found in attempt {current_round}. "
                        f"Waiting {sleep_time} seconds before trying again."
                    )
                    current_round += 1
                    time.sleep(sleep_time)
            # Write the Zarr table
            curr_table = ad.read_zarr(table_dict[table])
            write_table(
                new_image_group,
                table,
                curr_table,
                table_attrs=old_table_group.attrs.asdict(),
                overwrite=True,
            )

But actually, the ad.read_zarr sometimes hits the race condition now. I'll move this into the try/except part, that should hopefully fix this.