Improve interaction of BlobManager, GPT-4V with ADLS2

[pamelafox]

As discussed in meeting:

Currently, if you enable GPT-4-vision, it will upload images to the original Blob container. However, we should probably upload those images to ADLS2 container, if that's what the developer is using for their ListFileStrategy. We should also upload those images to the user's ADLS2 container for the upcoming user upload feature. This will probably easier to tackle when the upload feature is merged, so you can see all the situations.

This issue is for a: (mark with an x)

- [X] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)