[FEAT] Add MMAR/ MMAU-PRO for reasoning task support

[nhhoang96]

📌 Description

• Adding support for deep reasoning tasks, including MMAR and MMAU-PRO.
• Reusing llm-judge-binary for the MCQ evaluation for these two datasets.

🔗 Related Issue(s)

🛠️ Type of Change

• Bug fix (non-breaking change that fixes an issue)
• New feature (non-breaking change that adds functionality including new tasks)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Refactor / Code cleanup
• Maintenance / Chore / Task
• Other (please describe):

✅ How Has This Been Tested?

• Unit tests
• Integration tests
• Manual testing

Test Results / Screenshots (if applicable):

Sample run_config:

task_metric: 
  - ['mmar', 'llm_judge_binary']
  - ['mmau-pro', 'llm_judge_binary']
aggregate:
  - ['llm_judge_binary', ['mmau-pro']]
  - ['llm_judge_binary', ['mmar']]

Sample Experimental Log:
2025-09-20_23-40-59_544502_default.log

📸 Screenshots / Demos

N/A

📋 Checklist

• Code follows project style guidelines
• Tests have been added/updated (if applicable)
• Documentation has been updated (if applicable)
• Linked relevant issue(s)
• Self-reviewed my code

🙌 Additional Notes

N/A