Fix mirrored image export with CGBitmapContext renderer

Draw base CGImage without coordinate flip, apply flip+scale only for
annotations. Save via direct NSBitmapImageRep access instead of
tiffRepresentation to avoid transform issues. Also redirect clipboard
images to saveFolder instead of temp directory.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Sources/Editor/AnnotationRenderer.swift

@@ -26,21 +26,39 @@ enum AnnotationRenderer {
     }
 
     static func render(annotations: [AnnotationItem], onto image: NSImage) -> NSImage {
+        guard let srcCG = image.cgImage(forProposedRect: nil, context: nil, hints: nil) else { return image }
+        let w = srcCG.width
+        let h = srcCG.height
         let size = image.size
-        let result = NSImage(size: size)
-        result.lockFocusFlipped(true)
 
-        image.draw(in: NSRect(origin: .zero, size: size),
-                   from: .zero, operation: .sourceOver, fraction: 1,
-                   respectFlipped: true, hints: nil)
+        guard let cs = srcCG.colorSpace ?? CGColorSpace(name: CGColorSpace.sRGB),
+              let ctx = CGContext(
+                data: nil, width: w, height: h,
+                bitsPerComponent: 8, bytesPerRow: 0, space: cs,
+                bitmapInfo: CGImageAlphaInfo.premultipliedLast.rawValue
+              ) else { return image }
 
-        if let context = NSGraphicsContext.current?.cgContext {
-            for item in annotations {
-                draw(item, in: context)
-            }
+        ctx.draw(srcCG, in: CGRect(x: 0, y: 0, width: w, height: h))
+
+        ctx.translateBy(x: 0, y: CGFloat(h))
+        let sx = CGFloat(w) / size.width
+        let sy = CGFloat(h) / size.height
+        ctx.scaleBy(x: sx, y: -sy)
+
+        NSGraphicsContext.saveGraphicsState()
+        NSGraphicsContext.current = NSGraphicsContext(cgContext: ctx, flipped: true)
+
+        for item in annotations {
+            draw(item, in: ctx)
         }
 
-        result.unlockFocus()
+        NSGraphicsContext.restoreGraphicsState()
+
+        guard let resultCG = ctx.makeImage() else { return image }
+        let bitmapRep = NSBitmapImageRep(cgImage: resultCG)
+        bitmapRep.size = size
+        let result = NSImage(size: size)
+        result.addRepresentation(bitmapRep)
         return result
     }