feat: add query_transformer parameter to BaseRAGQuestionAnswerer

[srishti-1935]

Closes #67

What this PR does

Adds a query_transformer parameter to BaseRAGQuestionAnswerer.__init__()
that allows users to set query transformation behaviour before retrieval.

Changes

• Added query_transformer parameter (defaults to None to skip transformation)
• Stored as self.query_transformer in the class
• Applied transformation in answer_query() before retrieval if not None

Usage example

from pathway.xpacks.llm.prompts import prompt_query_rewrite
rag = BaseRAGQuestionAnswerer(
    llm=chat,
    indexer=vector_server,
    query_transformer=prompt_query_rewrite,
)

[CLAassistant]

All committers have signed the CLA.

[szymondudycz]

This PR doesn't do what it's supposed to do. Query transformer prompts need to be sent to LLM to be rewritten. The result of query transformer should only be used for the sake of document retrieval, answering should be based on the original quesion.

[szymondudycz]

pathway.xpacks.llm.prompts.prompt_query_rewrite and `pathway.xpacks.llm.prompts.prompt_query_rewrite_hyde`` are just prompts to be sent to an llm to ask to rewrite a question for the sake of retrieval. Replacing a question to RAG with these prompt makes the RAG completely wrong