support shared parameter in summary

[szha]

Description

print out number of parameters that are shared, in block summary.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage:
• Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
• Code is well-documented:
• Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
• To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

• print shared parameter count in summary

[szha]

For network like this:

net2 = nn.Sequential()
with net2.name_scope():
    net2.add(nn.Embedding(40, 30))
    net2.add(gluon.rnn.LSTM(30))
    net2.add(nn.Dense(40, flatten=False, params=net2[0].params))
net2.initialize()
net2.summary(mx.nd.ones((80, 32)))

the printed summary looks like:

--------------------------------------------------------------------------------
        Layer (type)                                Output Shape         Param #
================================================================================
               Input                                    (80, 32)               0
         Embedding-1                                (80, 32, 30)            1200
              LSTM-2                                (80, 32, 30)            7440
             Dense-3                                (80, 32, 40)            1240
================================================================================
Total params: 9880
Trainable params: 9880
Non-trainable params: 0
Shared params: 1200          # added in this PR
--------------------------------------------------------------------------------