Correct handling of seeds

[pgleeson]

@apdavison, I've updated brunel08.py to the latest version from the main PyNN repo & tested with the latest NEST (& NEURON, but see NeuralEnsemble/PyNN#382).

Regarding seeds though there are 2 mentioned:

# seed for random generator used when building connections
connectseed = 12345789

# seed for random generator(s) used during simulation
kernelseed  = 43210987

Only the latter is used:

rng = NumpyRNG(kernelseed, parallel_safe=True)

Should this be connectseed instead? Also, is there a way to hand the other seed to the simulator to ensure replicability?