[Enhancement] Metrics changes and Eval demo tasks addition

[NiraliPopat]

Summary

This PR is adding demo tasks for evaluation and the corresponding changes to support that.

Explain the features implemented:

• question answering eval task added at tasks/eval/question_answering/simpleqa
• classification eval task added at tasks/eval/classification/simpleqa
• precision, recall, and f1_score to get the average as well as per-class score
• unit_metric registry added
• fixes to support eval tasks

Performance impact (if any):

• N/A

How to Test the feature

Steps for reviewers to verify functionality:

1. run the task tasks/eval/question_answering/simpleqa
2. Observe tasks/eval/question_answering/simpleqa/MetricCollatorPostProcessor_.json file with results
3. run the task tasks/eval/classification/simpleqa
4. Observe tasks/eval/classification/simpleqa/MetricCollatorPostProcessor_.json file with results

Screenshots (if applicable)

NA

Checklist

• Lint fixes and unit testing done
• End to end task testing
• Documentation updated

[bidyapati-p]

We need to update documentation, this is first task on sygra platform. Lets discuss and update document as well as part of this

[bidyapati-p]

@NiraliPopat
Can you please attach the result file to review the output schema