Generalize and cleanup index-specific data processors and summary generators

[shreyah963]

Is your feature request related to a problem? Please describe

With the conversational agent now handling index routing, data_processors.py and summary_generators.py are the last per-index custom code. Adding a new index requires these file changes which can be cleaned up and made more generic.

Describe the solution you'd like

1. Add deduplication to query-time
2. Keep existing extract/summary functions as optional post-processors registered per index pattern, not hardcoded in if/elif branches
3. Add instructions to the conversational agent prompt to handle deduplication and summarization hints (e.g., "collapse by component, keep most recent by build_start_time")

Describe alternatives you've considered

No response

Additional context

No response

[krisfreedain]

@peterzhuamazon @DivyamSengar @getsaurabh02 @gaiksaya - can you take a look?

Catch All Triage - 1 2