Vector database with embeddings of select Bioconductor content

[slobentanzer]

In combination with #185, #186, and #187, we want to establish RAG processes based on free-form text gathered from Bioconductor resources, e.g., the vignettes and the support forum. Building a database with metadata that corresponds to the information in the aforementioned knowledge graphs will enable the LLM support agent to refer to specific documentation about software and troubleshooting when facing user questions about these issues.