Exercise 1: A Real update() Method — Possible Solution
====================================================================

THE METHOD
------------------------------
    def update(self, conn):
        non_id_fields = {name: col for name, col in self._fields.items() if name != 'id'}
        set_clause = ', '.join(f"{col} = ?" for col in non_id_fields.values())
        values = [getattr(self, name) for name in non_id_fields] + [self.id]
        sql = f"UPDATE {self._table} SET {set_clause} WHERE id = ?"
        conn.execute(sql, values)
        conn.commit()

VERIFIED, REAL RESULT
------------------------------
    a = Author.get(conn, 1)
    print(a.id, a.name)         # 1 Ann

    a.name = 'Annika'
    a.update(conn)

    refetched = Author.get(conn, 1)
    print(refetched.id, refetched.name)   # 1 Annika

    other = Author.get(conn, 2)
    print(other.id, other.name)           # 2 Bo  -- untouched

WHY THIS WORKS AS AN ANSWER
------------------------------
It builds a real SQL UPDATE ... SET ... WHERE id = ? statement rather
than reusing save()'s own INSERT OR REPLACE (which would re-insert the
entire row rather than modifying it in place, and would silently
rely on every column already being set correctly rather than changing
only what's genuinely different). Excluding 'id' from the SET clause
specifically avoids the pointless (and in a table with foreign-key
references, potentially dangerous) act of updating a row's own primary
key to itself. Re-fetching row 1 afterward confirms the real change
persisted through a genuine round trip to the database, and checking
row 2 separately confirms the WHERE id = ? clause correctly scoped the
update to exactly the one row intended.