Implement a pytest plugin to help with testing

[matthewelwell]

As part of some work in the core Flagsmith repository, we found that there isn't currently a good, generic solution for mocking the flagsmith client to return a given value for a flag at a given time. The ideal implementation as I see it would be to build a fixture that can be used as a context manager, or a function that looks something like this:

def my_test_function(set_flags: FlagsmithSetFlagsFixture) -> None:
    with override_flagsmith(flags=[Flag(name="foo", enabled=True, value="foobar")]):
        # in here, any calls to Flags.is_feature_enabled,Flags.get_flag and Flags.get_feature_value 
        # for the feature "foo" will return the values provided as an argument to override_flagsmith
        ...

def my_other_test_function(set_flags: FlagsmithSetFlagsFixture) -> None:
    set_flags(flags=[Flag(name="foo", enabled=True, value="foobar")])
    
    # For the remainder of this test, any calls to Flags.is_feature_enabled,Flags.get_flag 
    # and Flags.get_feature_value for the feature "foo" will return the values provided as 
    # an argument to override_flagsmith

See PR here for more context / discussions