multi-gpu tensorboard handlers initialization

https://github.com/Project-MONAI/model-zoo/blob/cf5e0322ee25b178b6cf841f3bd81e0a8adf2b16/models/spleen_ct_segmentation/configs/multi_gpu_train.json#L18

the multi-gpu override essentially set the trainer handlers to $@train#handlers[:-2] for the worker nodes. but because of the @train#handlers reference, the config parser will still trigger handler constructor calls on all nodes.

for tensorboard handlers this will be an issue, as each constructor call will create a new event log file. as a result the multinode log will have unnecessary event logging files. https://github.com/Project-MONAI/MONAI/blob/e36982b87bf87fb9559fc4d124e132b67f177d23/monai/handlers/tensorboard_handlers.py#L52-L55

Start with models/spleen_ct_segmentation/configs/multi_gpu_train.json at line 18 and trace how the @train#handlers reference is parsed for worker nodes. Read MONAI's tensorboard_handlers.py around lines 52-55 to understand why constructor calls create event logs. Done means multinode training no longer creates unnecessary TensorBoard event log files on worker nodes.