Converter improvements

[hunse]

Some improvements to get the Keras converter working with my network.

• Do not sort source_tensor list. This is no longer required, and is problematic for models that have an output that is used other places in the model (since the sorting puts all outputs at the end).
• Use a single bias connection in inference_only mode. We don't need to worry about proper bias sharing since we're not training them. It's much faster to use one connection to do all the biases in this case.

TODO:

• [x] Add a test of a network that has a middle layer as an output

[drasmuss]

Just a heads up, you might want to start from this branch https://github.com/nengo/nengo-dl/pull/134, as it has some of the changes you're adding already. It also has a more efficient convolutional bias implementation (I forgot that wasn't in master yet), which would be good to benchmark against the inference_only version to see if there is still an advantage.

[hunse]

I did a quick check of the new bias code in #130 on my network, and it appears to be as fast. I've put my old work in https://github.com/nengo/nengo-dl/tree/converter-hunse-inf-only-bias just in case we want it later, and rebased this onto #134.