[Bug]: HybridSearch crashes on empty results from OceanBase

[longbingljw]

Describe the bug

OceanBase returns NULL for empty results. pyobvector doesn't handle this.
When DBMS_HYBRID_SEARCH.SEARCH returns NULL (which happens when there are no matching results), HybridSearch.search() attempts to call json.loads(None), causing a TypeError.

Environment

0.2.18

Fast reproduce steps

from pyobvector.client.hybrid_search import HybridSearch

Connect to OceanBase

hs = HybridSearch(
uri="127.0.0.1:2881",
user="root@test",
password="password",
db_name="test"
)

result = hs.search(
index="test_table",
body={"query": {"match": {"content": "nonexistent_xyz123"}}}
)

Expected behavior

pyobvector handles empty results when they are NULL.

Actual behavior

TypeError: the JSON object must be str, bytes or bytearray, not NoneType

Additional context

No response