LibGPU+LibSoftGPU: Move size and pixel format information to GPU::Image

Size and format information are the same for every implementation and do
not need to be virtual. This removes the need to reimplement them for
each driver.

Userland/Libraries/LibGPU/Image.cpp

@@ -0,0 +1,32 @@
+/*
+ * Copyright (c) 2022, Stephan Unverwerth <s.unverwerth@serenityos.org>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <AK/Math.h>
+#include <LibGPU/Image.h>
+
+namespace GPU {
+
+Image::Image(void const* ownership_token, GPU::PixelFormat const& pixel_format, u32 width, u32 height, u32 depth, u32 max_levels)
+    : m_ownership_token { ownership_token }
+    , m_pixel_format { pixel_format }
+{
+    VERIFY(width > 0);
+    VERIFY(height > 0);
+    VERIFY(depth > 0);
+    VERIFY(max_levels > 0);
+
+    u32 number_of_levels_in_full_chain = max(max(AK::log2(width), AK::log2(height)), AK::log2(depth)) + 1;
+    m_mipmap_sizes.resize(min(max_levels, number_of_levels_in_full_chain));
+
+    for (u32 level = 0; level < m_mipmap_sizes.size(); ++level) {
+        m_mipmap_sizes[level] = { width, height, depth };
+        width = max(width / 2, 1);
+        height = max(height / 2, 1);
+        depth = max(depth / 2, 1);
+    }
+}
+
+}