maple_tree: add end of node tracking to the maple state

Analysis of the mas_for_each() iteration showed that there is a
significant time spent finding the end of a node.  This time can be
greatly reduced if the end of the node is cached in the maple state.  Care
must be taken to update & invalidate as necessary.

Link: https://lkml.kernel.org/r/20231101171629.3612299-5-Liam.Howlett@oracle.com
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Peng Zhang <zhangpeng.00@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/linux/maple_tree.h

@@ -393,6 +393,7 @@ struct ma_state {
 	unsigned char depth;		/* depth of tree descent during write */
 	unsigned char offset;
 	unsigned char mas_flags;
+	unsigned char end;		/* The end of the node */
 };
 
 struct ma_wr_state {