Clarify the relationship between acquisition.json and instrument.json

In the 2/3/26 behavior working group meeting, Bruno asked for clarification on the relationship between the acquisition and instrument JSON files.

When I looked [here](https://github.com/AllenNeuralDynamics/aind-software-docs/blob/main/docs/source/acquire_upload/acquire_data.md), it's true that there is no mention of how these files relate and if/how this relationship is validated.

We should add a section that discusses the requirement for active devices in the acquisition to match names of devices listed in the instrument. We should also specify when in the process this is validated and what the consequences will be if this validation fails (i.e., is upload prevented from rig to VAST? Is upload prevented from VAST to S3? How is the user notified? How should they fix it? etc)