To address the disconnection between day-ahead pricing and intra-day fault recovery, as well as the difficulty of coordinating benefits among multiple stakeholders, this paper proposes a day-ahead and intra-day multi-scale optimization method for distribution network-microgrid coordination under both normal and fault operating conditions. In the day-ahead stage, an improved Stackelberg game bi-level optimization model is established to describe the dynamic pricing and response behaviors between the distribution network and microgrids. Individual rationality and revenue fairness constraints are introduced to improve the participation willingness of microgrids and alleviate the possible market power abuse of the distribution network. Based on the Karush-Kuhn-Tucker optimality conditions and strong duality theory, the original non-convex bi-level model is transformed into a mixed-integer second-order cone programming model. In the intra-day stage, a rolling multi-objective optimization model with forest topology constraints is developed for operation correction under source-load fluctuations and fault conditions. A scenario-adaptive dynamic weighting ideal point method is used to aggregate and balance network loss, voltage deviation, and load shedding. Case studies on the modified IEEE 33-bus system show that the proposed method can improve the benefits of distribution network-microgrid coordination, support peak shaving and valley filling, and enhance the self-healing resilience of the system under extreme faults.