As [itex]\theta[/itex] goes from each point moves through a circle around the z-axis.
I don't know what more you want.
Unless you are using the "physics" convention rather than the "mathmatics" convention. Mathematics has [itex]\theta[/itex] measuring the angle the line from the origin to the given point's projection in the xy-plane makes with the positive x-axis while [itex]\phi[/itex] measures the angle the line from the origin to the point itself makes with the z-axis. The "physics" convetion swaps x and y. If that is what you mean then as [itex]\theta[/itex] goes from 0 to [itex]\pi[/itex], a point with the same r and [itex]\phi[/itex] is swung down through a semi-circle, with center at the origin and radius z, from (0, 0, z) to (0, 0, -z). If you then continue to increase [itex]\theta[/itex] from [itex]\pi[/itex] to [itex]2\pi[/itex] the point swings through the other haf of the same circle, from (0, 0, -z) back up to (0, 0, z). Yes, if we allowed [itex]\theta[/itex] to go from [itex]0[/itex] to [itex]2\pi[/itex] we could have the same point with two different sets of coordinates. [itex](r, \phi, \theta+ \pi)[/itex] would be the same point as [itex](r, \phi+ \pi, \theta)[/itex]. Which is why we restrict [itex]\theta[/itex] to be from 0 to [itex]\pi[/itex] rather than [itex]2\pi[/itex].