Add existing indices as a source data option; improve config autofilling

[ohltyler]

Description

This PR adds a third option to specify the source data - from an existing index. When selected, x number of documents are automatically fetched and populated. Additionally, when data is fetched from an existing index (or from an uploaded document, attempt to prefill some config data. More specifically:

• if index data is automatically populated, 1/ clear and update ML ingest processors (if applicable) with new text/vector fields (image is ignored for now, as we don't have a clear default pattern in ML processors for this use case. Additionally, would increase the level of fine-grained assumptions we already make)
• if uploaded document data is automatically populated, clear some fields in the ML ingest processors (if applicable)

Additional improvements:

• cleans up remaining stringify operations to use customStringify()
• moves index fetching to the top-level WorkflowDetail page so it is optimized and only called once
• moves remaining form components that should be compressed (EuiSuperSelect -> EuiCompressedSuperSelect)

Demo video, showing a workflow with some sample data, then creating another workflow, building a semantic search use case on top of that existing workflow's data using the new multi-option UI for data source selection. Also shows the underlying configs updated when data is populated via manual entry or file upload. Note this does not currently overwrite any default index mappings; users will still need to ensure this is up-to-date before creating.

screen-capture.13.webm

Issues Resolved

[List any issues this PR will resolve]

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.