Misreported cycle number in `gradients` file

[Andrew-S-Rosen]

Describe the bug The gradient file generated with --grad always reports that the results are for cycle 1 even when they are (presumably) for a different cycle.

The top of the gradient file in the example below is as follows:

$grad
  cycle =      1    SCF energy =   -26.43824279253   |dE/dxyz| =  0.000389

Note how it says cycle 1. The SCF energy, based on the log report below, seems to be for the final calculation in the set of cycles and not the first.

To Reproduce xtb coord.xyz --opt --grad > output.log

Output files below:

samples.zip

Expected behaviour The correct cycle number should be reported.

Additional context None.

[awvwgk]

The cycle reported here is a dummy rather than the actual cycle number.

[Andrew-S-Rosen]

Yup. I just thought it could be potentially confusing to the user.