Support relevant form toggling in advanced transform modals

[ohltyler]

Description

This PR adds specific, relevant forms to be exposed in the advanced transform modals, such that users can toggle and test directly within the modal, instead of navigating to/from the advanced settings fields. Specifically:

• exposes one_to_one toggling on INPUT transform for SEARCH RESPONSE only. This changes what the source input will be (single value vs. array of values)
• exposes full_response_path toggling on OUTPUT transforms for INGEST / SEARCH RESPONSE only. This changes what the source output will be (nested model response vs. full model response)

More details

• adds visual support for the default many-to-one behavior on input transform (when one_to_one is false). In this scenario, we collapse the values from all of the documents into a single list, to pass as a single parameter for one inference call via the ML processor
• updates generateTransform to support the above use case ^
• exposes override field for search resp processor
• limits input docs to the input transform, to be 10
• resets form inputs when form values are toggled

Demo video, showing one_to_one toggling on the input transform, and full_response_path toggling on the output transform. Note the model input selector in the output field changes from a selector when false, to a freeform text when true. This is because when it is true, the full model response path is passed, and users must write custom JSON path to parse out the relevant values they need.

screen-capture.24.webm

Issues Resolved

Resolves #367

Check List

• Commits are signed per the DCO using --signoff

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[saimedhi]

Code Changes Looks good to me. From the UI, one_to_one toggling on the input transform, and full_response_path toggling on the output transform seem to work rightly. Thank you, Tyler.