feat: change default fuzziness from AUTO to 0 in OpenSearch retrievers

fuzziness="AUTO" was causing severe performance regressions (30s → 6s
search times) for large indexes. Switching the default to 0 (exact
matching) avoids the clause explosion that AUTO triggers on big corpora.
Users who need fuzzy matching can still pass fuzziness="AUTO" explicitly.

Closes #3187

Author: SyedShahmeerAli12

integrations/opensearch/src/haystack_integrations/components/retrievers/opensearch/bm25_retriever.py

@@ -29,7 +29,7 @@ def __init__(
         *,
         document_store: OpenSearchDocumentStore,
         filters: dict[str, Any] | None = None,
-        fuzziness: int | str = "AUTO",
+        fuzziness: int | str = 0,
         top_k: int = 10,
         scale_score: bool = False,
         all_terms_must_match: bool = False,

integrations/opensearch/src/haystack_integrations/components/retrievers/opensearch/open_search_hybrid_retriever.py

@@ -93,7 +93,7 @@ def __init__(
         embedder: TextEmbedder,
         # OpenSearchBM25Retriever
         filters_bm25: dict[str, Any] | None = None,
-        fuzziness: int | str = "AUTO",
+        fuzziness: int | str = 0,
         top_k_bm25: int = 10,
         scale_score: bool = False,
         all_terms_must_match: bool = False,

integrations/opensearch/src/haystack_integrations/document_stores/opensearch/document_store.py

@@ -1057,7 +1057,7 @@ def _bm25_retrieval(
         query: str,
         *,
         filters: dict[str, Any] | None = None,
-        fuzziness: int | str = "AUTO",
+        fuzziness: int | str = 0,
         top_k: int = 10,
         scale_score: bool = False,
         all_terms_must_match: bool = False,
@@ -1115,7 +1115,7 @@ async def _bm25_retrieval_async(
         query: str,
         *,
         filters: dict[str, Any] | None = None,
-        fuzziness: str = "AUTO",
+        fuzziness: int | str = 0,
         top_k: int = 10,
         scale_score: bool = False,
         all_terms_must_match: bool = False,

integrations/opensearch/tests/test_bm25_retriever.py

@@ -68,7 +68,7 @@ def test_to_dict(_mock_opensearch_client):
                 "type": "haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore",
             },
             "filters": {},
-            "fuzziness": "AUTO",
+            "fuzziness": 0,
             "top_k": 10,
             "scale_score": False,
             "filter_policy": "replace",
@@ -163,7 +163,7 @@ def test_run():
     mock_store._bm25_retrieval.assert_called_once_with(
         query="some query",
         filters={},
-        fuzziness="AUTO",
+        fuzziness=0,
         top_k=10,
         scale_score=False,
         all_terms_must_match=False,
@@ -183,7 +183,7 @@ async def test_run_async():
     mock_store._bm25_retrieval_async.assert_called_once_with(
         query="some query",
         filters={},
-        fuzziness="AUTO",
+        fuzziness=0,
         top_k=10,
         scale_score=False,
         all_terms_must_match=False,
@@ -343,7 +343,7 @@ def test_run_with_runtime_document_store():
     runtime_store._bm25_retrieval.assert_called_once_with(
         query="some query",
         filters={},
-        fuzziness="AUTO",
+        fuzziness=0,
         top_k=10,
         scale_score=False,
         all_terms_must_match=False,
@@ -374,7 +374,7 @@ async def test_run_async_with_runtime_document_store():
     runtime_store._bm25_retrieval_async.assert_called_once_with(
         query="some query",
         filters={},
-        fuzziness="AUTO",
+        fuzziness=0,
         top_k=10,
         scale_score=False,
         all_terms_must_match=False,

integrations/opensearch/tests/test_open_search_hybrid_retriever.py

@@ -78,7 +78,7 @@ class TestOpenSearchHybridRetriever:
                 },
             },
             "filters_bm25": None,
-            "fuzziness": "AUTO",
+            "fuzziness": 0,
             "top_k_bm25": 10,
             "scale_score": False,
             "all_terms_must_match": False,
@@ -219,7 +219,7 @@ def test_run_with_extra_runtime_params(self, mock_embedder):
             filters={"key": "value"},
             top_k=1,
             all_terms_must_match=False,
-            fuzziness="AUTO",
+            fuzziness=0,
             scale_score=False,
             custom_query=None,
         )
@@ -252,7 +252,7 @@ def test_run_in_pipeline(self, mock_embedder):
             filters={"param_a": "default"},
             top_k=10,
             all_terms_must_match=False,
-            fuzziness="AUTO",
+            fuzziness=0,
             scale_score=False,
             custom_query=None,
         )