Consider using Document Intelligence method for extracting figure images

[pamelafox]

We currently use Python Pillow plus pymupdf, but apparently DI has a method as well:
https://learn.microsoft.com/en-us/python/api/overview/azure/ai-documentintelligence-readme?view=azure-python-preview#extract-figures-from-documents

We should try that and see if the results are the same (quality/latency/cost).
We could also look at the code to see if they basically do the same thing.

[pamelafox]

This was done in the new multimodal approach:
https://github.com/Azure-Samples/azure-search-openai-demo/releases/tag/2025-08-29