[Feature Proposal] Build Search Focused Sample Application

[macohen]

Summary

In order to show demonstrable progress in search features, we want to have a sample index and a search application that gets updated with each feature we release. The sample application is visible by default when starting up OpenSearch and/or OpenSearch Dashboards. When I talk about "search" in this case, I mean non-logging/observability use cases. Cases like e-commerce search, document archive search, and website search.

Questions

1. What dataset can be used that is interesting enough to expose most or all search related features, static enough so it doesn't need to be updated every release, and inclusive enough to not require specific domain knowledge?
2. Would it be useful to only include a search index to start as an initial release and add UI related features later?