Train.A3c: Use the A3C algorithm to train a model
In MartinKies/USLR: Reinforcement Learning with R

Returns a list as given by Initialise.A3C with stored experiences for each worker.

1	Train.A3c(model.par, algo.par, algo.var, game.object, episodes.max)

`model.par`	A list as given by `Get.Def.Par.Neural.Network.A3C` or `Get.Def.Par.Neural.Network.A3C.LSTM`.
`algo.par`	A list as given by `Get.Def.Par.A3C`.
`algo.var`	A list as given by `Initialise.A3C`.
`game.object`	A Game Object (list) as defined by `Get.Game.Object.<NAME>`.
`episodes.max`	Specifies how many episodes should be played in total.