Add robust field filtering in TBEDataConfig.from_json (#5164)

Summary:
X-link: facebookresearch/FBGEMM#2163


Add field validation and filtering to `TBEDataConfig.from_json` to handle JSON inputs, used in benchmark, that may contain unknown or extra fields.

The current `tbe_reporter` has adhoc properties added to the output that mismatch with the current structure.

This change makes JSON deserialization more robust by filtering out unrecognized fields and logging warnings for missing expected fields and info messages for ignored extra fields. This allows the code to gracefully handle evolving schema or data from external sources without breaking deserialization.

Differential Revision: D87658217

Author: gchalump

fbgemm_gpu/fbgemm_gpu/tbe/bench/tbe_data_config.py

@@ -9,6 +9,7 @@
 
 import dataclasses
 import json
+import logging
 from typing import Any, Optional
 
 import torch
@@ -78,7 +79,20 @@ def from_dict(cls, data: dict[str, Any]):
     @classmethod
     # pyre-ignore [3]
     def from_json(cls, data: str):
-        return cls.from_dict(json.loads(data))
+        raw = json.loads(data)
+        allowed = {f.name for f in dataclasses.fields(cls)}
+        filtered = {k: v for k, v in raw.items() if k in allowed}
+        missing = allowed - set(filtered.keys())
+        extra = set(raw.keys()) - allowed
+        if missing:
+            logging.warning(
+                f"TBEDataConfig.from_json: Missing expected fields not loaded: {sorted(missing)}"
+            )
+        if extra:
+            logging.info(
+                f"TBEDataConfig.from_json: Ignored unknown fields from input: {sorted(extra)}"
+            )
+        return cls.from_dict(filtered)
 
     def dict(self) -> dict[str, Any]:
         tmp = dataclasses.asdict(self)