Issue #, if available:

Description of changes:

• Update results and task definitions in benchmarks/ to v0.6.0.
• Allow setting n_resamples=None to disable bootstrap CIs in leaderboard and pairwise_comparison.
• Handle deprecated fields in results when loading evaluation summaries.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.