feat(suite benchmarking): add progress tracking - #1498

Description

Adds run_suite_with_progress() function to provide real-time progress feedback when benchmarking models across OpenML suites.

Motivation

When running benchmarks on large suites (e.g., OpenML-CC18), users currently have no visibility into progress, making it difficult to:

• Monitor long-running experiments
• Estimate completion time
• Debug stuck/slow tasks

This is particularly important for the AI-on-Demand Platform where researchers run multi-hour benchmarks.

Related Issue

#1497

Changes

• New function: openml.study.run_suite_with_progress()
• Progress bar with tqdm showing completion, ETA, and speed
• Full backwards compatibility (no breaking changes)
• Comprehensive tests (3 test cases)
• Tutorial with examples

Example Usage

from sklearn.ensemble import RandomForestClassifier
from openml.study import run_suite_with_progress

clf = RandomForestClassifier(n_estimators=10, random_state=42)
runs = run_suite_with_progress(suite_id=334, model=clf)

Output

Benchmarking OpenML-Tiny: 100%|████████████| 7/7 [01:35<00:00, 13.6s/task]