Exercise 2: Isolating State Across Three Tests — Possible Solution
====================================================================

THE FIX
------------------------------
    def test_first_order_is_order_number_one():
        processed_orders = []          # fresh, owned by this test alone
        result = process_order('A1', 50, processed_orders)
        return result['total_processed'] == 1

    (identical pattern applied to the second and third tests, each with
    its own fresh processed_orders list)

RESULTS (all 6 possible orderings of 3 tests)
------------------------------
    ('first', 'second', 'third'):  all pass: True
    ('first', 'third', 'second'):  all pass: True
    ('second', 'first', 'third'):  all pass: True
    ('second', 'third', 'first'):  all pass: True
    ('third', 'first', 'second'):  all pass: True
    ('third', 'second', 'first'):  all pass: True

    ALL 6 orderings fully green: True

Every one of the 3 tests passes in every one of the 6 possible
orderings - a complete elimination of the order dependency, not just
an improvement to it.

WHY THIS WORKS AS AN ANSWER
------------------------------
This confirms the chapter's own fix scales the same way the bug did:
with each test creating its own processed_orders list, no test can
ever observe another test's own effects, so "which test ran first" has
become a question with no bearing on any test's own result. The fix
generalizes to any number of tests for the same reason the bug did -
isolation (or its absence) is a property of the shared resource, not
of how many tests happen to be sharing it.