Tikhonov regularization continuation methods and the trust-region updating strategy for linearly equality-constrained optimization problems.