Make an option to get embedding from the already indexed content to avoid going to LLM

Problem/Motivation

Currently the view mode is used to render related content. This is nice to curate what to use, but in case that curated content is the same as what is used for Search API Indexing, its doing an extra LLM call.

Steps to reproduce

No option to use existing embedding

Proposed resolution

1. Make option to use existing embedding
2. Maybe set as default

Remaining tasks

MR

User interface changes

New option

API changes

N/A

Data model changes

N/A