Which Team Has the Widest Pay Gap?

An external pay-equity audit has landed at Kestrel Labs, and the first thing the auditors want is a crude but revealing figure: for each team, the distance between its best-paid and its worst-paid person. A wide distance is not proof of anything on its own, but it tells the auditors which teams to open up first.

The awkward part is that the answer for a team depends on two different rows of the employee table at once — and which two is not known in advance. Marketing is a further wrinkle: it has exactly one employee, so its best-paid and worst-paid person are the same human being. The audit still expects Marketing to appear on the report rather than quietly falling off it.

employees — one row per person. department_id is the team they sit on. salary is annual pay in dollars.

The table already exists in the database — there is nothing to create or load.

Task: Write a query returning two columns, department_id and salary_spread, with one row per department that has at least one employee. salary_spread is the highest salary on that team minus the lowest, in whole dollars. A team of one produces a row with a spread of 0, not an absent row. A department with no employees at all produces no row. Rows come back sorted by department_id, smallest first.

Example output — shape only, on an invented company; team 6 is a team of one, which is why it reads 0 rather than being missing.