[v2] [Feature] Support for customLabels in the evaluation output images

[Udayraj123]

Note: this enhancement is to be built on top of the dev branch as the feature is not available in master branch currently. It is part of OMRChecker v2

Is your feature request related to a problem? Please describe.
We now support outputting evaluation summary on the marked bubble images(reference).

But currently it is limited to separate field labels, and ignores the customLabels provided in the template.json. One common use for customLabels is the two-digit integer type question, where we can show both marked bubbles as green if the multi-column answer was correct.

Describe the solution you'd like

• Need to find a way to visually display the correct/incorrect verdicts without an ambiguity in case of the customLabels where multiple fields are concatenated before running evaluation.
• Should document all possible cases of verdicts when using custom labels
• Add a test sample containing these different cases (you can also reuse any existing sample that contains customLabels)

Describe alternatives you've considered
N/A

Additional context
Here's the todo comment corresponding to above enhancement

# linked_custom_labels = [custom_label if (field_label in field_labels) else None for (custom_label, field_labels) in template.custom_labels.items()]
# is_part_of_custom_label = len(linked_custom_labels) > 0
# TODO: replicate verdict: question_has_verdict = len([if field_label in questions_meta else None for field_label in linked_custom_labels])

Note: please share your queries and approaches on discord for quicker discussions:

• Link to the Discord invite
• Link to the Hacktoberfest 2024 channel