Issue with an interval without any data.

[hgolsen]

There is an issues where the results of an interval is without any data, and with a rather cryptic error message (length=204 ms, required length 204 ms).

[hgolsen]

We have checked in a fix for this now. There was a problem reestimating sn-ratio for an interval and Crava continued with erroneous value, resulting with inf-values in the result. This is now fixed. We will use a the globally estimated sn-ratio if there is not enough data to estimate for one single interval.

For the cryptic error message we actually require more data (larger than 204 ms). This is updated to say the correct required amount of data (here 205 ms).