Some fixes in Python fuzzer logic

[tochilinak]

Description

Changes:

• New logic for functions with default arguments: first try to fuzz only positional arguments, then all of them.
• Added new Python tree node: FakeNode (instead of Java null). If fuzzer provides object with this node, we don't start execution with this data.
• BaselineAlgorithm now provides random type each 6 iterations.
• Logic for abstract classes is now the same as for protocols.

How to test

Manual tests

Default arguments: https://github.com/TheAlgorithms/Python/blob/master/machine_learning/gradient_descent.py (function _error).

Random type: https://github.com/TheAlgorithms/Python/blob/master/data_structures/queue/circular_queue.py (function enqueue of class CircularQueue). Mypy does not pass any types for data parameter, bacause it consideres self.array to be list[None].

Test for abstract class:

from typing import AbstractSet

def id_(x: AbstractSet[int]):
    return x

Generated output:

import sys
sys.path.append(r'../../../utbot-python/samples/easy_samples')
import builtins
import general
import unittest


class TestTopLevelFunctions(unittest.TestCase):
    # region Test suites for executable general.id_
    
    # region FUZZER
    
    def test_id_(self):
        """
        x = builtins.set[typing.Any]
        """
        actual = general.id_(set())
        
        self.assertEqual(set(), actual)
    # endregion
    
    # endregion

Self-check list

Check off the item if the statement is true. Hint: [x] is a marked item.

Please do not delete the list or its items.

• I've set the proper labels for my PR (at least, for category and component).
• PR title and description are clear and intelligible.
• I've added enough comments to my code, particularly in hard-to-understand areas.
• The functionality I've repaired, changed or added is covered with automated tests.
• Manual tests have been provided optionally.
• The documentation for the functionality I've been working on is up-to-date.