Fix tag_feas code injection in retrieval ranking (#13923)

## Summary
- remove eval-based parsing from retrieval rank feature scoring
- validate `tag_feas` at write time in chunk APIs and SDK routes
- add regression tests for safe parsing and malicious payload rejection

## Details
`tag_feas` is intended to be structured rank-feature data, but the
retrieval ranking path was evaluating stored values as Python
expressions. This change treats `tag_feas` strictly as data.

### What changed
- replace `eval()` in `rag/nlp/search.py` with safe parsing via
`json.loads()` and optional `ast.literal_eval()` compatibility for
legacy Python-dict strings
- strictly filter parsed values down to `dict[str, finite number]`
- reject invalid `tag_feas` payloads at write time in web chunk routes
and SDK document chunk routes
- add focused regression tests to prove executable strings are ignored
and invalid payloads are rejected

## Validation
- `python -m pytest test/unit_test/common/test_tag_feature_utils.py
test/unit_test/rag/test_rank_feature_scores.py -q`

---------

Co-authored-by: unknown <zhenglinkai@CCN.Local>
Co-authored-by: Yingfeng Zhang <yingfeng.zhang@gmail.com>

api/apps/chunk_app.py

@@ -38,6 +38,7 @@ from api.utils.api_utils import (
     get_request_json,
 )
 from common.misc_utils import thread_pool_exec
+from common.tag_feature_utils import validate_tag_features
 from rag.app.qa import beAdoc, rmPrefix
 from rag.app.tag import label_question
 from rag.nlp import rag_tokenizer, search
@@ -161,7 +162,10 @@ async def set():
             return get_data_error_result(message="`tag_kwd` must be a list of strings")
         d["tag_kwd"] = req["tag_kwd"]
     if "tag_feas" in req:
-        d["tag_feas"] = req["tag_feas"]
+        try:
+            d["tag_feas"] = validate_tag_features(req["tag_feas"])
+        except ValueError as exc:
+            return get_data_error_result(message=f"`tag_feas` {exc}")
     if "available_int" in req:
         d["available_int"] = req["available_int"]
 
@@ -328,7 +332,10 @@ async def create():
             return get_data_error_result(message="`tag_kwd` must be a list of strings")
         d["tag_kwd"] = req["tag_kwd"]
     if "tag_feas" in req:
-        d["tag_feas"] = req["tag_feas"]
+        try:
+            d["tag_feas"] = validate_tag_features(req["tag_feas"])
+        except ValueError as exc:
+            return get_data_error_result(message=f"`tag_feas` {exc}")
     image_base64 = req.get("image_base64", None)
 
     try:

api/apps/sdk/doc.py

@@ -39,6 +39,7 @@ from common.constants import FileSource, LLMType, ParserType, RetCode, TaskStatu
 from common.metadata_utils import convert_conditions, meta_filter
 from common.misc_utils import thread_pool_exec
 from common.string_utils import remove_redundant_spaces
+from common.tag_feature_utils import validate_tag_features
 from rag.app.qa import beAdoc, rmPrefix
 from rag.app.tag import label_question
 from rag.nlp import rag_tokenizer, search
@@ -963,7 +964,10 @@ async def add_chunk(tenant_id, dataset_id, document_id):
             return get_error_data_result("`tag_kwd` must be a list of strings")
         d["tag_kwd"] = req["tag_kwd"]
     if "tag_feas" in req:
-        d["tag_feas"] = req["tag_feas"]
+        try:
+            d["tag_feas"] = validate_tag_features(req["tag_feas"])
+        except ValueError as exc:
+            return get_error_data_result(f"`tag_feas` {exc}")
     import base64
 
     image_base64 = req.get("image_base64", None)
@@ -1202,7 +1206,10 @@ async def update_chunk(tenant_id, dataset_id, document_id, chunk_id):
             return get_error_data_result("`tag_kwd` must be a list of strings")
         d["tag_kwd"] = req["tag_kwd"]
     if "tag_feas" in req:
-        d["tag_feas"] = req["tag_feas"]
+        try:
+            d["tag_feas"] = validate_tag_features(req["tag_feas"])
+        except ValueError as exc:
+            return get_error_data_result(f"`tag_feas` {exc}")
     tenant_embd_id = DocumentService.get_tenant_embd_id(document_id)
     if tenant_embd_id:
         model_config = get_model_config_by_id(tenant_embd_id)