related to issue #170

Hi,
I have implemented an initial version of power profiling. It was based on the torchinfo (the display style is the same as torchinfo, however, it reports the number of synapses, neurons, spiking and all events with some power estimations, simillarly to KerasSpiking ). Currently, I implemented :

• 'estimate_energy' function which takes a nn.Module, some extra information like input tensor on which we want to run the network or input shape and information about the Device, and similarly to KerasSpiking, it produces some estimation about the energy required for inference [snntorch/energy_estimation/estimate_energy.py ]
• A basic model of the device contained in DeviceProfile (similar to Keras Spiking, currently considers only energy per neuron event, energy per synapse event and is the device neuromorphic) and a way to register new devices (DeviceProfileRegistry) [snntorch/energy_estimation/device_profile.py and snntorch/energy_estimation/device_profile_registry.py]
• only support for Linear layers, Convolutional layers, Activation functions and any nested layers containing these (calculating synapse/neuron counts and events for them). I added a simple way to register new types of layers: for a given layer type (identified by class name), I can add some function that will count the number of synapses/neurons (given the layer information and input/output tensors), another function which will count the number of total and spiking events(given the layer information and input/output tensors), and optional function which will override default way of calculating energy, and return a custom energy term [snntorch/energy_estimation/layer_parameter_event_calculator.py]

To see how in current shape it is used, and what it looks like please see the tests/test_energy_estimation.py file.

Please let me know if you think this implementation is good enough and what I can improve to make it part of the repository.
Best Regards