migration: detect mislabeled SO arms and relabel to 'unknown'

A robot_type of so100/so101 is treated as wrong when the joint dim isn't a
multiple of 6, or (when names are present) the first 6 joints don't match the
canonical SO set. Such datasets are migrated structurally to v3.0 with joints
left untouched and robot_type relabeled 'unknown', instead of being skipped or
degrees-converted on a false assumption.

community_v3_migration/classify.py

@@ -25,6 +25,9 @@ def is_so_robot_type(rt: str) -> bool:
     return bool(rt) and (rt.startswith(SO_PREFIXES) or rt in SO_EXACT) and rt not in NEVER_FIX
 
 
+SO_JOINTS = ("shoulder_pan", "shoulder_lift", "elbow_flex", "wrist_flex", "wrist_roll", "gripper")
+
+
 def is_end_effector(info: dict) -> bool:
     """True if action/observation.state are task-space end-effector features (e.g. ``ee_x``,
     ``ee_roll``) rather than joint angles. Such datasets are out of scope for the joint fix."""
@@ -38,6 +41,22 @@ def is_end_effector(info: dict) -> bool:
     return False
 
 
+def is_mislabeled_so(info: dict) -> bool:
+    """True when ``robot_type`` claims SO but the features prove it isn't a standard 6-DOF SO arm:
+    the joint dim isn't a multiple of 6, or (when names are present) the first 6 joints don't match
+    the canonical SO set. Such datasets keep their joints untouched and are relabeled 'unknown'
+    rather than being degrees-converted on a wrong assumption."""
+    feats = info.get("features", {})
+    dims = [feats[c]["shape"][0] for c in ("action", "observation.state") if feats.get(c, {}).get("shape")]
+    if any(d % 6 != 0 for d in dims):
+        return True
+    for key in ("action", "observation.state"):
+        names = [str(n).lower() for n in (feats.get(key, {}).get("names") or [])]
+        if len(names) >= 6 and not all(SO_JOINTS[i] in names[i] for i in range(6)):
+            return True
+    return False
+
+
 def load_info(root: Path) -> dict:
     return json.loads((Path(root) / "meta" / "info.json").read_text())
 
@@ -106,6 +125,10 @@ def classify(root) -> dict:
         return {**out, "is_so": False, "encoding": "end_effector",
                 "note": "task-space end-effector features"}
 
+    if is_so_robot_type(rt) and is_mislabeled_so(info):
+        return {**out, "is_so": False, "encoding": "non_so", "mislabeled_so": True,
+                "note": "robot_type claims SO but joint dim/names don't match a 6-DOF SO arm"}
+
     is_so = is_so_robot_type(rt)
     if not is_so:
         return {**out, "is_so": False, "encoding": "non_so"}