Add dataset tags to SDK for identification (DE-7033)

Expose dataset tags through the Python SDK so customers can identify
datasets labeled by Scale vs other vendors via the API.

- Add `tags` field to DatasetInfo model (returned by dataset.info())
- Add get_tags(), add_tags(), remove_tags() methods to Dataset class
- Use POST /tags/remove instead of DELETE to avoid proxy body-stripping
- Use pydantic v1/v2 compat shim for null-coercion validator
- Guard against passing a bare string instead of a list

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: vinay553

nucleus/data_transfer_object/dataset_info.py

@@ -1,4 +1,12 @@
-from typing import Any, Dict, List, Optional
+from typing import TYPE_CHECKING, Any, Dict, List, Optional
+
+if TYPE_CHECKING:
+    from pydantic.v1 import validator
+else:
+    try:
+        from pydantic.v1 import validator
+    except ImportError:
+        from pydantic import validator
 
 from nucleus.pydantic_base import DictCompatibleModel
 
@@ -14,6 +22,7 @@ class DatasetInfo(DictCompatibleModel):
         slice_ids: List :class:`Slice` IDs associated with the :class:`Dataset`
         annotation_metadata_schema: Dict defining annotation-level metadata schema.
         item_metadata_schema: Dict defining item metadata schema.
+        tags: List of tags associated with the :class:`Dataset`.
     """
 
     dataset_id: str
@@ -24,3 +33,8 @@ class DatasetInfo(DictCompatibleModel):
     # TODO: Expand the following into pydantic models to formalize schema
     annotation_metadata_schema: Optional[Dict[str, Any]] = None
     item_metadata_schema: Optional[Dict[str, Any]] = None
+    tags: List[str] = []
+
+    @validator("tags", pre=True, always=True)  # pylint: disable=used-before-assignment
+    def coerce_null_tags(cls, v):  # pylint: disable=no-self-argument
+        return v if v is not None else []

nucleus/dataset.py

@@ -432,6 +432,49 @@ def info(self) -> DatasetInfo:
         dataset_info = DatasetInfo.parse_obj(response)
         return dataset_info
 
+    def get_tags(self) -> List[str]:
+        """Fetches tags associated with the dataset.
+
+        Returns:
+            List of tag strings associated with this dataset.
+        """
+        response = self._client.make_request(
+            {}, f"dataset/{self.id}/tags", requests.get
+        )
+        return response["tags"]
+
+    def add_tags(self, tags: List[str]) -> List[str]:
+        """Adds tags to the dataset.
+
+        Args:
+            tags: List of tag strings to add.
+
+        Returns:
+            Updated list of all tags on the dataset.
+        """
+        if isinstance(tags, str):
+            raise TypeError("tags must be a list of strings, not a single string")
+        response = self._client.make_request(
+            {"tags": tags}, f"dataset/{self.id}/tags", requests.post
+        )
+        return response["tags"]
+
+    def remove_tags(self, tags: List[str]) -> List[str]:
+        """Removes tags from the dataset.
+
+        Args:
+            tags: List of tag strings to remove.
+
+        Returns:
+            Updated list of remaining tags on the dataset.
+        """
+        if isinstance(tags, str):
+            raise TypeError("tags must be a list of strings, not a single string")
+        response = self._client.make_request(
+            {"tags": tags}, f"dataset/{self.id}/tags", requests.delete
+        )
+        return response["tags"]
+
     @deprecated(
         "Model runs have been deprecated and will be removed. Use a Model instead"
     )