ADD: ocr options inputs for all ocr engines + fix for model training options and wait submission

[benleetownsend]

Summary

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist:

• Jira story/task is put into PR title
• Code has been self-reviewed
• Hard-to-understand areas have been commented
• Documentation has been updated
• My changes generate no new warnings
• Unit tests have been added for base functionality and known edge cases
• Any dependent changes have been merged and published in downstream modules

---

Note

Medium Risk
Touches multiple client-facing request builders and expands OCR configuration surfaces; risk is mainly API/serialization compatibility and potential breaking behavior for callers relying on prior parameter types or implicit pagination.

Overview
Extends dataset OCR configuration support by adding new OCR option input types (including READAPI_TABLES_V2) and plumbing them through CreateDataset/CreateEmptyDataset, with stricter validation to ensure only one OCR options block is provided.

Improves a few client request ergonomics/bugs: ListGallery now accepts structured filter dicts, UpdateModelGroupSettings correctly serializes model_training_options to JSON, and WaitForSubmissions accepts a single submission id and sets an explicit limit to avoid over-fetching.

^{Written by Cursor Bugbot for commit 452a612. This will update automatically on new commits. Configure here.}

[jacobmanderson]

maybe we can have a base ocr options class that different versions can subclass

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}