Migrate the MVP into kotaemon (#108)

- Migrate the MVP into kotaemon.
- Preliminary include the pipeline within chatbot interface.
- Organize MVP as an application.

Todo:

- Add an info panel to view the planning of agents -> Fix streaming agents' output.

Resolve: #60
Resolve: #61 
Resolve: #62

knowledgehub/indices/qa/text_based.py

@@ -1,6 +1,6 @@
 import os
 
-from kotaemon.base import BaseComponent, Document, RetrievedDocument
+from kotaemon.base import BaseComponent, Document, Node, RetrievedDocument
 from kotaemon.llms import AzureChatOpenAI, BaseLLM, PromptTemplate
 
 from .citation import CitationPipeline
@@ -21,6 +21,9 @@ class CitationQAPipeline(BaseComponent):
         temperature=0,
         request_timeout=60,
     )
+    citation_pipeline: CitationPipeline = Node(
+        default_callback=lambda self: CitationPipeline(llm=self.llm)
+    )
 
     def _format_doc_text(self, text: str) -> str:
         """Format the text of each document"""
@@ -52,9 +55,7 @@ class CitationQAPipeline(BaseComponent):
         self.log_progress(".prompt", prompt=prompt)
         answer_text = self.llm(prompt).text
         if use_citation:
-            # run citation pipeline
-            citation_pipeline = CitationPipeline(llm=self.llm)
-            citation = citation_pipeline(context=context, question=question)
+            citation = self.citation_pipeline(context=context, question=question)
         else:
             citation = None