Add model wrapper modules (embedding, reranker, llm, preference_extractor)

Add Python wrappers for:
- Qwen3/Nemotron embedding models
- BGE/Qwen3 rerankers
- vLLM/Llama/Qwen LLM backends
- GPT-4o/LLM-based preference extractors

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

1 files changed, 17 insertions, 0 deletions

diff --git a/src/personalization/models/preference_extractor/base.py b/src/personalization/models/preference_extractor/base.py
new file mode 100644
index 0000000..850292f
--- /dev/null
+++ b/src/personalization/models/preference_extractor/base.py
@@ -0,0 +1,17 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+from typing import Any, Dict, List
+from personalization.retrieval.preference_store.schemas import ChatTurn, PreferenceList
+
+class PreferenceExtractorBase(ABC):
+    @abstractmethod
+    def extract_turn(self, turns: List[ChatTurn]) -> PreferenceList:
+        """
+        Extract preferences from a window of chat turns (history + current query).
+        """
+        raise NotImplementedError
+
+# Alias for backward compatibility if needed, 
+# though specific extractors should inherit from PreferenceExtractorBase now.
+PreferenceExtractor = PreferenceExtractorBase