Custom PdfLoader implementation support

[pkt1583]

Currently there is only pypdfLoader supported (

rag-experiment-accelerator/rag_experiment_accelerator/doc_loader/pdfLoader.py

Line 64 in a85beac

loader = PyPDFLoader(file_path=file_path)

). It would be good to have ability to inject different pdfloader implementation. Any implementation that supports langchain document interface should be able to be plugged in form the config while running the experiment.

This would help to quickly try out with various pdf loader available either at https://python.langchain.com/docs/modules/data_connection/document_loaders/pdf/ or a custom created one