CPF Gets Stuck in While Loop with Adaptive Stepping Enabled Due to Tiny Step Sizes

When running Continuation Power Flow (CPF) with adaptive stepping enabled 'cpf.adapt_step' = 1, the algorithm can get stuck in an infinite loop inside the main while ~done.flag loop. This occurs without triggering any rollback mechanisms. The root cause appears to be the adaptive step size shrinking to extremely small values (e.g., 1e-7 or smaller), causing the next predicted state to be effectively identical to the current state. As a result, the continuation parameter (lam) doesn't change, while event detection doesn't trigger terminations.

I reported the outputs when I met this problem as follows. Hope the author can fix this issue. Thank you very much!

step 235 : stepsize = 0.2 lambda = 4.293 2 corrector Newton steps
step 236 : stepsize = 0.2 lambda = 4.330 2 corrector Newton steps
step 237 : stepsize = 0.2 lambda = 4.367 2 corrector Newton steps
step 238 : stepsize = 0.2 lambda = 4.402 2 corrector Newton steps
step 239 : stepsize = 0.2 lambda = 4.676 5 corrector Newton steps
step 240 : stepsize = 0.0604 lambda = 4.676 1 corrector Newton steps
step 241a : stepsize = 0.121 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 241 : stepsize = 0.0967 lambda = 4.677 2 corrector Newton steps
step 242a : stepsize = 0.0183 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 242b : stepsize = 0.0143 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 242c : stepsize = 0.0112 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 242d : stepsize = 0.00884 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 242e : stepsize = 0.00701 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 242 : stepsize = 0.00558 lambda = 4.677 1 corrector Newton steps
step 243a : stepsize = 0.00114 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 243b : stepsize = 0.000902 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 243c : stepsize = 0.000717 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 243 : stepsize = 0.00057 lambda = 4.677 1 corrector Newton steps
step 244a : stepsize = 0.000117 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 244 : stepsize = 9.29e-05 lambda = 4.677 0 corrector Newton steps
step 245a : stepsize = 1.9e-05 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245b : stepsize = 1.51e-05 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245c : stepsize = 1.2e-05 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245d : stepsize = 9.57e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245e : stepsize = 7.6e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245f : stepsize = 6.04e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245g : stepsize = 4.8e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245h : stepsize = 3.82e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245i : stepsize = 3.04e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245j : stepsize = 2.41e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245k : stepsize = 1.92e-06 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 245 : stepsize = 1.52e-06 lambda = 4.677 0 corrector Newton steps
step 246a : stepsize = 3.13e-07 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 246 : stepsize = 2.49e-07 lambda = 4.677 0 corrector Newton steps
step 247a : stepsize = 5.1e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247b : stepsize = 4.05e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247c : stepsize = 3.22e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247d : stepsize = 2.56e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247e : stepsize = 2.04e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247f : stepsize = 1.62e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247g : stepsize = 1.29e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247h : stepsize = 1.02e-08 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247i : stepsize = 8.12e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247j : stepsize = 6.46e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247k : stepsize = 5.13e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 247 : stepsize = 4.08e-09 lambda = 4.677 0 corrector Newton steps
step 248 : stepsize = 8.37e-10 lambda = 4.677 0 corrector Newton steps
step 249a : stepsize = 3.42e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249b : stepsize = 2.71e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249c : stepsize = 2.16e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249d : stepsize = 1.72e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249e : stepsize = 1.36e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249f : stepsize = 1.08e-09 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249g : stepsize = 8.61e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249h : stepsize = 6.85e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249i : stepsize = 5.44e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249j : stepsize = 4.33e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249k : stepsize = 3.44e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249l : stepsize = 2.73e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249m : stepsize = 2.17e-10 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 249 : stepsize = 1.73e-10 lambda = 4.677 0 corrector Newton steps
step 250a : stepsize = 3.54e-11 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 250b : stepsize = 2.82e-11 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 250 : stepsize = 2.24e-11 lambda = 4.677 0 corrector Newton steps
step 251a : stepsize = 4.59e-12 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 251 : stepsize = 3.65e-12 lambda = 4.677 0 corrector Newton steps
step 252a : stepsize = 7.49e-13 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 252b : stepsize = 5.95e-13 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
step 252c : stepsize = 4.73e-13 lambda = 4.679 3 corrector Newton steps ^ ROLLBACK
...
step 1552 : stepsize = 1.4e-15 lambda = 4.677 0 corrector Newton steps
step 1553 : stepsize = 1.4e-15 lambda = 4.677 0 corrector Newton steps
step 1554 : stepsize = 1.4e-15 lambda = 4.677 0 corrector Newton steps
step 1555 : stepsize = 1.4e-15 lambda = 4.677 0 corrector Newton steps

Start by tracing the main while ~done.flag loop with 'cpf.adapt_step' = 1 enabled, focusing on how shrinking step sizes affect lam and event detection. Reproduce the reported progression toward a step size near 1e-15, then ensure the CPF run cannot remain in the loop when the predicted state and continuation parameter stop changing.