include model fixed parameters in sample_lhs result

[saschaishikawa]

@dhadka: I've been using fixed_parameters to pass in Model parameters that are constant (but may change between experiments). For instance,

model.fixed_parameters.update({
    "constant_param1": 2045, 
    "constant_param2": 20
})

However, I run into problems evaluating the SOWs generated by sample_lhs(), since fixed_parameters aren't included in it and the model still expects these as parameters. I figured instead of running this each time:

SOWs = sample_lhs(model, 10)
results1 = evaluate(model, update(SOWs, {"constant_param1": 2045, "constant_param2":20}) )
results2 = evaluate(model, update(SOWs, {"constant_param1": 2050, "constant_param2":15}) )

could we have sample_lhs automatically pass along the constants in the SOWs? Are there cases when we wouldn't want this to happen?