Optimize ignore_list to use set and single iteration

This pr addressed the comment by Umago, to iterate docs
list once as opposed to twice. His suggestion also also
optimises for O(1) membership testing instead of list.

Changes:
 - Convert ignore_list to ignore_set for faster lookups
 - Single loop builds both docs_to_check and ignored_docs

Author: malingatembo

src/lightspeed_rag_content/document_processor.py

@@ -692,12 +692,13 @@ def process(
         if unreachable_action != "warn":
             # Separate docs into those we should check and those in ignore_list
             if ignore_list:
-                docs_to_check = [
-                    doc for doc in docs if doc.metadata.get("title") not in ignore_list
-                ]
-                ignored_docs = [
-                    doc for doc in docs if doc.metadata.get("title") in ignore_list
-                ]
+                docs_to_check = []
+                ignored_docs = []
+                for doc in docs:
+                    if doc.metadata.get("title") in ignore_list:
+                        ignored_docs.append(doc)
+                    else:
+                        docs_to_check.append(doc)
             else:
                 docs_to_check = docs
                 ignored_docs = []