assertions on ai output, not vibes

Build a small eval runner. The starter has a list of cases (each with name, input, and expected) and a stub classify(text) that returns "yes" or "no" based on whether the text contains "cancel". Write a function run_suite(cases) that:

• Calls classify(case["input"]) for each case.
• Compares the lowercased actual to the lowercased expected.
• Returns a tuple (passed, total).

Then call run_suite(cases) and print the result as passed/total.

Expected output:

3/3