From Indeterminacy to Determinacy: Augmenting Logical Reasoning Capabilities with Large Language Models