Learners/feature_synthesis.py: GenerateDerivedFeatures generates a list of tuples of Z3 values now, need to make it return a list of feature vectors.

*** Make a base Scorer class that can extended***: the scorer base class can be extended by classes such as EntropyScore, MaxConjunctScore

Data/feature_vector.py: Why range(len(values) - 1) before?
Before, values contains test label at the end, so the actual values don't include the last one. Now we have the constructor with additional field testLabel, so it should be range(len(values)) now.

Teacher/houdini.py: f.varZ3[i] out of range
It comes from 1, if the input values are all actual values (without testLabel) are we use range(len(values) - 1), then we will miss the value at the end. If we change it to range(len(values)) we should be good.

Data/feature_vector.py: valuesZ3 should be False but print is True
Note!!! In z3py, BoolVal('False') returns True, but BoolVal(False) returns False. We should initialize with bool value rather than string.

Leaner/houdini.py: substitute() doesn't work. The reason is that before PrecisFeature is initialized with string of variable name, which makes no sense for derived features such as New_Count != Old_Count. Now, we added new field isDerived in constructor of PrecisFeature, when isDerived is True, we set PrecisFeature with a Z3 expression direcetly, if not then we use string of variable name and string of variable type to initialize the feature.