[test] Mark scale-offset tests as XFAIL for image, xlib-fallback.

The scale-offset tests trigger a loss-of-precision bug when
converting double matrices from cairo to pixman format.

http://lists.cairographics.org/archives/cairo/2008-November/015671.html

test/Makefile.am

@@ -1071,6 +1071,8 @@ $(REFERENCE_IMAGES)
 #                         until they have come up with an interface and
 #                         semantics that actually work. :-)
 # long-lines            - range overflow of fixed-point
+# scale-offset-image    - loss of precision converting a cairo matrix to
+# scale-offset-similar    pixman's fixed point format.
 # self-copy-overlap     - vector surfaces take snapshot of patterns in contrast
 #                         to the raster backends which don't. One solution
 #                         would be to clone overlapping areas of dst/source, so

test/scale-offset-image.c

@@ -135,7 +135,7 @@ draw (cairo_t *cr, int width, int height)
 
 CAIRO_TEST (scale_offset_image,
 	    "Tests drawing surfaces under various scales and transforms",
-	    "XFAIL=!ps surface scale-offset", /* keywords */
+	    "XFAIL=!ps,image,xlib-fallback surface scale-offset", /* keywords */
 	    NULL, /* requirements */
 	    320, 320,
 	    NULL, draw)

test/scale-offset-similar.c

@@ -136,7 +136,7 @@ draw (cairo_t *cr, int width, int height)
 
 CAIRO_TEST (scale_offset_similar,
 	    "Tests drawing surfaces under various scales and transforms",
-	    "XFAIL=!ps surface scale-offset", /* keywords */
+	    "XFAIL=!ps,image,xlib-fallback surface scale-offset", /* keywords */
 	    NULL, /* requirements */
 	    320, 320,
 	    NULL, draw)