Untracking variables created with lambda functions in the logger does not actually work

[tkevinbest]

Specifically, creating a lambda func in the untrack method will not work, though we show it in the docstring:

        Examples:
            # Single variable untracking
            >>> class MyClass:
            ...     def __init__(self):
            ...         self.value = 42
            >>> obj = MyClass()
            >>> LOGGER.track_variable(lambda: obj.value, "value")
            >>> LOGGER.update()
            >>> LOGGER.flush_buffer()
            >>> LOGGER.untrack_variable(lambda: obj.value)

            # Multiple variable untracking
            >>> class MyClass:
            ...     def __init__(self):
            ...         self.value1 = 42
            ...         self.value2 = 84
            >>> obj = MyClass()
            >>> LOGGER.track_variable(
            ...     [lambda: obj.value1, lambda: obj.value2],
            ...     ["value1", "value2"]
            ... )
            >>> LOGGER.update()
            >>> LOGGER.flush_buffer()
            >>> LOGGER.untrack_variable([lambda: obj.value1, lambda: obj.value2])

Instead of the lambda paradigm, I think we need to internally create these lambda functions so we can keep track of their ID. I'll implement if you agree.

Originally posted by @tkevinbest in #336