Make page_in_from_vnode 2x faster.

...by adding a new class called Ext2Inode that inherits CoreInode.
The idea is that a vnode will wrap a CoreInode rather than InodeIdentifier.
Each CoreInode subclass can keep whatever caches they like.

Right now, Ext2Inode caches the list of block indices since it can be very
expensive to retrieve.

VirtualFileSystem/VirtualFileSystem.cpp

@@ -54,6 +54,8 @@ auto VirtualFileSystem::makeNode(InodeIdentifier inode) -> RetainPtr<Node>
     if (!metadata.isValid())
         return nullptr;
 
+    auto core_inode = inode.fileSystem()->get_inode(inode);
+
     InterruptDisabler disabler;
 
     CharacterDevice* characterDevice = nullptr;
@@ -74,6 +76,7 @@ auto VirtualFileSystem::makeNode(InodeIdentifier inode) -> RetainPtr<Node>
     fileSystem->retain();
 
     vnode->inode = inode;
+    vnode->m_core_inode = move(core_inode);
     vnode->m_cachedMetadata = { };
 
 #ifdef VFS_DEBUG