feat(search): enhance phrase matching and reranking logic with improved async handling

tests/ebook_search/test_rerank.py

@@ -10,7 +10,7 @@ import pytest
 
 from python.ebook_search.config import EbookSearchConfig, RerankConfig, load_rerank_config
 from python.ebook_search.rerank import rerank_chunks
-from python.ebook_search.search import SearchResult, apply_rerank, skip_rerank
+from python.ebook_search.search import SearchResult, apply_rerank
 
 if TYPE_CHECKING:
     from pytest_mock import MockerFixture
@@ -57,10 +57,10 @@ def test_config_defaults_enable_reranking(mocker: MockerFixture) -> None:
     assert config.timeout_seconds == 30
 
 
-def test_reranking_disabled_returns_original_fused_order() -> None:
+async def test_reranking_disabled_returns_original_fused_order(mocker: MockerFixture) -> None:
     config = EbookSearchConfig(rerank=RerankConfig(enabled=False), top_k=2)
 
-    response = skip_rerank("query", candidates(), config)
+    response = await apply_rerank(mocker.Mock(), "query", candidates(), config, rerank=False)
 
     assert response.rank_label == "Hybrid"
     assert [result.chunk_id for result in response.results] == [1, 2]
@@ -133,7 +133,7 @@ async def test_vllm_rerank_timeout_raises(mocker: MockerFixture) -> None:
     config = EbookSearchConfig(rerank=RerankConfig(enabled=True), top_k=2)
 
     with pytest.raises(httpx.TimeoutException, match="timeout"):
-        await apply_rerank(mocker.Mock(), "query", candidates(), config)
+        await apply_rerank(mocker.Mock(), "query", candidates(), config, rerank=True)
 
 
 async def test_malformed_vllm_rerank_json_does_not_crash_search(mocker: MockerFixture) -> None: