Arc end_angle should be renamed to span_angle

[BillyDM]

After scratching my head for a while, I found out that the end_angle field in iced::widget::canvas::path::arc::Arc is not the angle where the end of the arc lies. It is actually the span of the arc.

In other words, the end of the arc lies at start_angle + end_angle. The field end_angle should be renamed to something like span_angle to prevent confusion.

https://docs.rs/iced/0.1.1/iced/widget/canvas/path/struct.Arc.html

[hecrj]

We should keep the name unchanged and fix the Frame implementation instead.

[Vanille-N]

If I may, I believe it's as simple as

// graphics/src/widget/canvas/path/builder.rs
@@ -84,7 +84,7 @@ impl Builder
- sweep_angle: math::Angle::radians(arc.end_angle),
+ sweep_angle: math::Angle::radians(arc.end_angle - arc.start_angle),

I've checked that all other uses of end_angle (2 in total) are correct:

// graphics/src/widget/canvas/path/builder.rs
impl From<Arc> for Elliptical {
    fn from(arc: Arc) -> Elliptical {
        Elliptical {
            // fields omitted
            end_angle: arc.end_angle,   // line 41
        }
    }
}

and miraculously, the one other place where end_angle is used also needs no change (which perhaps explains why it went unnoticed)

// graphics/src/widget/canvas/path/builder.rs
#[inline]
pub fn circle(&mut self, center: Point, radius: f32) {
    self.arc(Arc {
        center,
        radius,
        start_angle: 0.0,
        end_angle: 2.0 * std::f32::consts::PI   // line 157
    });
}

Seeing that this is labeled good first issue, I assume it's okay for a beginner like me to open a pull request ?

[hecrj]

@Vanille-N Sounds good! By all means, feel free to open a PR.