Fix ptflops module registration

Fixed

• FLOPs module registration compatibility with ptflops >=v0.6.8.

Upgraded Lambda

Added

• Call-mode specific functions in co.Lambda

Update Lambda and auto convertible modules

Added

• Support for functor in co.Lambda

Remove

• nn.LayerNorm from automatically convertible modules

Update auto convertible nn modules

Added nn.LayerNorm to automatically convertible modules

`ParallelDispatch` module, updated prints and minor bugfixes

Added

• ParallelDispatch module.
• Conditional predicate print in __repr__.

Fixed

• Sequential padding computation.
• Lambda __repr__ function prints.

Removed

• CI testing for python v3.6.

Receptive field, Phantom padding, Reshape, Lambda update and various fixes

Added

• Added phantom_padding to Residual.
• Added receptive_field property.
• Added Reshape module.

Changed

• Rename forward_shrink argument to auto_shrink in Delay.
• Torch requirement to v1.9.
• Replace Lambda unsqueeze_step with takes_time and new default to False.

Fixed

• padding property in sequence.
• delay property in sequence.
• strict mode in load_state_dict.

Removed

• Assertion error in BroadcastReduce for modules with different delays.

Add `forward_shrink` option to `Delay` and `Residual`

Added

• Add forward_shrink option to Delay and Residual.

GPU support and Constant modules

Added

• Add Constant.
• Add Zero.
• Add One.

Fixed

• co.ConvXd cuda compatibility.

Add `flatten_state_dict` state variable.

Added

• Add flatten_state_dict state variable.

Removed

• Debug message for Convolutions with padding.

Fix `call_mode` for `co.Linear`

Fixed

• call_mode for Linear.