[Typing][C-12] Add type annotations for `python/paddle/distributed/co…

…mmunication/all_gather.py` (PaddlePaddle#66051)

python/paddle/base/core.pyi

@@ -96,3 +96,9 @@ class iinfo:
 
 def is_compiled_with_cuda() -> bool: ...
 def set_nan_inf_debug_path(arg0: str) -> None: ...
+
+class task:
+    def is_completed(self) -> bool: ...
+    def is_sync(self) -> bool: ...
+    def synchronize(self) -> None: ...
+    def wait(self, timeout: int = 0) -> bool: ...

python/paddle/distributed/communication/all_gather.py

@@ -12,6 +12,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, TypeVar
+
 import numpy as np
 
 import paddle
@@ -23,8 +27,20 @@
     convert_tensor_to_object,
 )
 
+if TYPE_CHECKING:
+    from paddle import Tensor
+    from paddle.base.core import task
+    from paddle.distributed.communication.group import Group
+
+    _T = TypeVar("_T")
+
 
-def all_gather(tensor_list, tensor, group=None, sync_op=True):
+def all_gather(
+    tensor_list: list[Tensor],
+    tensor: Tensor,
+    group: Group | None = None,
+    sync_op: bool = True,
+) -> task | None:
     """
 
     Gather tensors from all participators and all get the result. As shown
@@ -42,7 +58,7 @@ def all_gather(tensor_list, tensor, group=None, sync_op=True):
             should be float16, float32, float64, int32, int64, int8, uint8, bool, bfloat16, complex64 or complex128.
         tensor (Tensor): The Tensor to send. Its data type
             should be float16, float32, float64, int32, int64, int8, uint8, bool, bfloat16, complex64 or complex128.
-        group (Group, optional): The group instance return by new_group or None for global default group.
+        group (Group|None, optional): The group instance return by new_group or None for global default group.
         sync_op (bool, optional): Whether this op is a sync op. The default value is True.
 
     Returns:
@@ -68,7 +84,9 @@ def all_gather(tensor_list, tensor, group=None, sync_op=True):
     return stream.all_gather(tensor_list, tensor, group, sync_op)
 
 
-def all_gather_object(object_list, obj, group=None):
+def all_gather_object(
+    object_list: list[_T], obj: _T, group: Group = None
+) -> None:
     """
 
     Gather picklable objects from all participators and all get the result. Similar to all_gather(), but python object can be passed in.