Skip tasks with insufficient samples for pass@k instead of raising ex…

[evtimovi]

When computing pass@k metrics with k > 1, tasks that have fewer than k samples are now gracefully skipped with a warning log message instead of raising a ValueError that would terminate the entire results processing.

Changes:

• Replace ValueError with warning log when n_samples < k for a task
• Add logging module import and logger instance
• Collect skipped groups and log them with full context (dataset, agent, attack, task_id, and sample count)
• Add check for empty DataFrame after filtering in aggregate_results
• Update docstrings to reflect new behavior (Note instead of Raises)
• Also includes refactoring: remove job_name from group_cols to allow aggregating across multiple runs of the same experiment
• Add generic variant_name support alongside legacy template_short_name

[dedeswim]

I fear this might lead to silently computing wrong metrics? Should we have a non-strict mode that behaves like this via a flag instead?

[meta-codesync]

@evtimovi has imported this pull request. If you are a Meta employee, you can view this in D92393526.