Relocate the old Decoder::new method to mod test

The old `Decoder::new()` method would not pull the image header out of
the iterator. This was fine for the design then, but I've just decided
that I want to collect this info and store it in the Decoder struct.

I've moved the old method down into the test module so that I can still
use it to test the decoder parts alone. Same philosophy as the
`new_with_previous_pixel()` and `new_with_backbuffer()` methods.

src/decoder.rs

@@ -89,20 +89,6 @@ where
             run_len: 0,
         })
     }
-
-    fn new(bytes: I) -> Self {
-        Self {
-            back_buffer: [PixelRGBA::zero(); 64],
-            prev_pixel: PixelRGBA {
-                r: 0,
-                g: 0,
-                b: 0,
-                a: 255,
-            },
-            bytes,
-            run_len: 0,
-        }
-    }
 }
 
 impl<'input, I> Iterator for Decoder<I>
@@ -230,6 +216,23 @@ mod test {
             }
         }
 
+        // The decoder includes image metadata, now, and construction attempts
+        // to extract it from the input iterator. For the decoder tests, this
+        // needs to be skipped. Thus, we get *another* magic constructor
+        // available only to the test module.
+        fn new_with_no_metadata(bytes: I) -> Self {
+            Self {
+                width: 0,
+                height: 0,
+                channels: 4,
+                colorspace: 0,
+                back_buffer: [PixelRGBA::zero(); 64],
+                prev_pixel: PixelRGBA { r: 0, g: 0, b: 0, a: 255 },
+                bytes,
+                run_len: 0,
+            }
+        }
+
         fn peek_prev_pixel(&self) -> &PixelRGBA {
             &self.prev_pixel
         }