Layer integration (#83)

* integrated parallel layers for ease of building models

* integrated 2.5d layers

* cleaned codes and unit tests

* added log metric by step hook; updated imagenet benchmark; fixed some bugs

* reworked initialization; cleaned codes

Co-authored-by: BoxiangW <45734921+BoxiangW@users.noreply.github.com>

examples/simclr_cifar10_data_parallel/train_simclr.py

@@ -41,9 +41,7 @@ def build_dataset_test():
     )
 
 def main():
-    colossalai.launch_from_torch(config='./config.py',
-                                 host='localhost',
-                                 port=29500)
+    colossalai.launch_from_torch(config='./config.py')
     
     # get logger
     logger = get_dist_logger()