This plugin hooks into the RabbitHole to ask the Language Model to summarize groups of text chunks.
Summaries are stored in the vector memory, along with the original chunks.
The summarization can help the Cat to have a more comprehensive understanding of the content of the documents.
Warning The summarization increases considerably the uploading time if the document size is big.
Moreover, when this plugin is enabled the Cat makes and additional call to the language model API.