Fix gizmo line width issue when using perspective (#9067)

- In bevy_polyline, we discovered an issue that happens when line width
is smaller than 1.0 and using perspective. It would sometimes end up
negative or NaN. I'm not entirely sure _why_ it happens.

- Make sure the width doesn't go below 0 before multiplying it with the
alpha

Here's a link to the bevy_polyline issue
ForesightMiningSoftwareCorporation/bevy_polyline#46

I'm not sure if the solution is correct but it solved the issue in my
testing.

Co-authored-by: François <mockersf@gmail.com>

crates/bevy_gizmos/src/lines.wgsl

@@ -64,7 +64,7 @@ fn vertex(vertex: VertexInput) -> VertexOutput {
 #endif
 
     // Line thinness fade from https://acegikmo.com/shapes/docs/#anti-aliasing
-    if line_width < 1. {
+    if line_width > 0.0 && line_width < 1. {
         color.a *= line_width;
         line_width = 1.;
     }
@@ -79,11 +79,11 @@ fn vertex(vertex: VertexInput) -> VertexOutput {
         let epsilon = 4.88e-04;
         // depth * (clip.w / depth)^-depth_bias. So that when -depth_bias is 1.0, this is equal to clip.w
         // and when equal to 0.0, it is exactly equal to depth.
-        // the epsilon is here to prevent the depth from exceeding clip.w when -depth_bias = 1.0 
-        // clip.w represents the near plane in homogenous clip space in bevy, having a depth
+        // the epsilon is here to prevent the depth from exceeding clip.w when -depth_bias = 1.0
+        // clip.w represents the near plane in homogeneous clip space in bevy, having a depth
         // of this value means nothing can be in front of this
-        // The reason this uses an exponential function is that it makes it much easier for the 
-        // user to chose a value that is convinient for them
+        // The reason this uses an exponential function is that it makes it much easier for the
+        // user to chose a value that is convenient for them
         depth = clip.z * exp2(-line_gizmo.depth_bias * log2(clip.w / clip.z - epsilon));
     }