Missing leaderboards for manual tasks

[Alexiland]

First of all, thanks for putting up this comprehensive benchmarks on BioML agents.

When I prepared the dataset, all manual tasks (OpenProblem) are having empty leaderboard.csv files due to

@register_data_source("manual")
class ManualDataSource(DataSource):
    ...
    def get_leaderboard(self, source_config: Dict[str, Any]) -> pd.DataFrame:
        """
        Manual tasks don't have external leaderboards.
        
        Args:
            source_config: Configuration (not used for manual tasks)
            
        Returns:
            Empty DataFrame with expected leaderboard columns
        """
        # Return empty leaderboard for manual tasks
        return pd.DataFrame(columns=['teamName', 'score', 'submissionDate']) 

I'm wondering if there's any instruction to to obtain the leaderboard info for manual tasks to reproduce the result in Table 4 of the preprint.