docs: add nn/distributed to index routines and helpers

Author: stefpi

docs/src/python/nn.rst

@@ -183,3 +183,4 @@ In detail:
    nn/functions
    nn/losses
    nn/init
+   nn/distributed

docs/src/python/nn/distributed.rst

@@ -0,0 +1,31 @@
+.. _distributed:
+
+Distributed
+-----------
+
+Helper Routines
+^^^^^^^^^^^^^^^
+
+The :code:`mlx.nn.layers.distributed` package contains helpful routines for creating sharded layers easily from existing :class:`Modules <mlx.nn.Module>`.
+
+.. currentmodule:: mlx.nn.layers.distributed
+.. autosummary::
+   :toctree: _autosummary
+
+   shard_linear
+   shard_inplace
+
+Layers
+^^^^^^
+
+The :code:`mlx.nn` package contains helpful layers for creating distributed models with sharded parameters.
+
+.. currentmodule:: mlx.nn
+.. autosummary::
+   :toctree: _autosummary
+   :template: nn-module-template.rst
+
+   AllToShardedLinear
+   ShardedToAllLinear
+   QuantizedAllToShardedLinear
+   QuantizedShardedToAllLinear