Possible refactor to make each statistical test an object/class.

tests_as_objs.py

# Idea: Make each statistical test its own class with an execute and visualization function
# For example: 

class Students_T():
  def execute(dataset, design, predictions, combined_data: CombinedData): 
    # perform statistical test
    pass
  def visualize():
    # add visualization code 
    pass
  def text_output(): 
    # create the text template here
    pass
  def __str___(): 
    # provide the text_output and visualization
    text_output()
    visualize()
    
# If we have this, then we could refactor __create_relate_vardata function (line 368, https://github.com/tea-lang-org/tea-lang/blob/master/tea/vardata_factory.py)
# One option: Keep everything the same until line 427 and do something like this: 
valid_tests = []
for test in tests: 
  test_obj = create_test_object(test) # some function like this that takes as input the list of tests that are valid (from the solver) and then instantates an object for each valid test
  valid_tests.append(test_obj)
  
for vt in valid_tests: 
  vt.execute(...) # pass all the needed parameters
  
 # handle multiple comparisons

return valid_tests # returning a list will likely break Tea V0, but would be fine for V1. Also, Tea V0 could work around returning a list for testing.