Add PDF source support to DeepSearch

Summary

DeepSearch currently does not support PDF indexing. To enable this, raw binary PDF files must be downloaded to a temporary location for interpretation and chunking.

Steps to Reproduce

Not provided.

Expected Behaviour

DeepSearch should support PDF sources by downloading raw binary to a temporary location, interpreting the content, and chunking it.

Actual Behaviour

DeepSearch does not support PDF indexing.

Environment

Not provided.

Reported by retoor via DevPlace.