strange eccentricity enhancement (bug)

Dear all,

I wanted to use the Soker eccentricity enhancement for my binary model and stumbled over a bug in the code:

In order to model it properly i set the initial eccentrictiy to 0.14

in the binary_controls I am using the options:

```
do_tidal_circ = .true.
!both stars are massive so used the "_rad" option
circ_type_1 = 'Hut_rad'
circ_type_2 = 'Hut_rad'
use_eccentricity_enhancement = .true.
!limit eccentricity so that we can see an increase (0 * something = 0!)
min_eccentricity = 0.001
max_eccentricity = 0.99

mdot_scheme = "Kolb"
```

So when starting the model something weird happens, either the eccentricity is directly at 0.99 or 0.001!
I started investigating it by copying the part of the code responsible for the eccentricity enhancement in
run_binary_extras.f90 and turning on

```
use_other_edot_enhance = .true.
```

it took me some while but I think the problem is with the value b% mdot_donor_theta which seems not to
be set at the beginning of the calculations as it gives very large (and random) values -1e292 or 1e292 (or
something in this range). This explains why after some starts the eccentricity was 0.99 and sometimes 0.001.

However, I don't know exactly when the mdot_donor_theta is calculated but I think it is after the eccentricity
enhancement is appied? I hope you can help me to solve this issue.

Btw. is there a possibility to calculate mdot_donor_theta for the 'contact' or 'roche_lobe' mdot_scheme, I guess
not as they are solved in a different way, but worth questioning if it would be in principle possible.

Sincerely,
Daniel Pauli