Add random_state to upliftTreeClassifier and upliftRandomForestClassifier

[flippercy]

So far the prediction of upliftTreeClassifier and upliftRandomForestClassifier is not repeatable, which is probably due to the missing of random_state. Could someone add this as a parameter, please?

Thank you.

[paullo0106]

You're right, thanks for bringing this up. It's currently not a reproducible result for the UpliftTree, #275 was a previous attempt by @RuslanTs to address the issue, as a reference

[manojbalaji1]

@paullo0106 I think we need to use randomstate rather than np.random.seed(42). I think there's this issue raised that np.random.seed is not thread-safe. I think that should be the issue. Let me investigate this problem and try to fix it

Reference: https://stackoverflow.com/questions/7029993/differences-between-numpy-random-and-random-random-in-python#:~:text=For%20numpy.random,to%20the%20contrary).

[jeongyoonlee]

Added in #422.