VillarFit fails with positive overflow

[hombit]

MCMC+lmsder, mcmc_niter=1<<13, lmsder_niter=20

called `Result::unwrap()` on an `Err` value: RangeError::PosOverflow(..)', /Users/hombit/.cargo/registry/src/github.aaakk.us.kg-1ecc6299db9ec823/light-curve-feature-0.4.1/src/features/villar_fit.rs:327:5

t = [ 740.3246  750.3043  760.4053  769.232   770.3784  789.3297  791.4037
  801.1434  812.3367  814.1552  815.2388  831.0803  875.9615  886.097
  894.0538  896.0607  897.0574  906.0077 1098.3772 1105.3289 1121.335
 1128.3058 1130.246  1140.3579 1149.3875 1153.2346 1157.1772 1170.1466
 1175.2944 1187.1001 1196.2018 1223.9636 1231.9731 1251.0542 1259.077
 1276.0114 1277.0122]
flux = [273.80392 240.27126 277.84772 315.173   322.88892 343.2325  328.50708
 380.0725  366.28265 377.44757 368.2116  382.0028  297.67258 291.70386
 323.76587 351.57016 368.64804 346.69186 335.55072 326.6749  331.6794
 368.32587 356.75455 398.51483 388.39102 408.93744 414.015   423.02805
 432.19604 386.59708 419.76703 347.92648 343.05618 329.53763 318.1823
 249.52344 252.61868]
fluxerr = [10.539378  16.701963  12.045661  13.417532  10.200134  10.353363
 12.642858  11.900937  12.752062  11.356     12.095391  12.088432
  8.629117  10.640575  16.658218  14.747993  16.652096  11.831199
  9.96417   10.513032  11.939197  11.237393  13.0192175 11.654888
 12.284968  10.904655   8.079373  12.025639  11.322596  13.843957
 13.0899935 15.582092   8.104286  14.444137   8.318043  12.236689
 18.179932 ]