Easy modification of the refocusing time bounds

[HKaras]

When using chirped pulses it is often helpful to increase the bound for the refocusing time when importing an experiment model into dl.dipolarmodel. Currently this is set +-50ns however when using chirped pulses that are over 100ns long this can be an issue.

In my personal fork I have modified the dl.dipolarmodel function to take an additional parameter where the +- uncertainty can be adjusted. I think this should become standard.

[luisfabib]

I agree, this should be an exposed options to the user.

A more intuitive way (for users) to implement this could be to add an optional keyword argument to the experiment model functions (e.g. ex_4pdeer) which allow them to specify the longest pulse length used in the experiment. From that value the program could automatically adjust the boundaries on the refocusing times, e.g. reftime +- 3*max(pulselength). So for example for standard 16ns rectangular pulses one would have reftime +- 48ns (similar to what is now hard-coded) and for shaped pulses of 100ns one would have reftime +- 300ns.

[HKaras]

I agree, this would be a nice way to implement this.