Rule evaluation optimization : alpha node ordering

Motivation: This Story aims at rule evaluation optimization so both Kogito and drools core engine usage will get benefits for overall rule execution. It focuses on Alpha Node Ordering to maximize sharing and minimize the number of evaluations. Other related JIRAs may be added later.

Goals: Better rule execution performance. For each improvement, add benchmark and measure the performance gain.

Impact: Better rule execution performance. It might have overhead during rule build time. Will confirm with benchmark anyway.