Add Loop Flatten and Unroll and Jam optimizations to Omax cfg

These optimizations see some good improvements and the downstream equivalents
are enabled for -Omax on Arm Compiler 6.

Author: stuij

Omax.cfg

@@ -6,3 +6,5 @@
 -mllvm -lsr-complexity-limit=1073741823 \
 -mllvm -force-attribute=main:norecurse \
 -mllvm -enable-dfa-jump-thread \
+-mllvm -enable-loop-flatten \
+-mllvm -enable-unroll-and-jam \