reconsider 20 minute correction in culm_addtime for calc_havengetallen

[veenstrajelmer]

reconsider 20 minute correction in culm_addtime, this seems odd since we relate it to the moon culmination that happens on the ocean. Compare time reference to Gele Boekje or 1991.0

Reference from Gemiddelde Getijkromme 1991.0:

Current settings are:

• timezone correction implemented in neater handling of time in calc_havengetallen #163 (and more manually before)
• • 4 * 12 hours and 25 minutes (including offset of 4 when relating extremes to moonculminations)
• • 20 minutes to correct from 0 to 5 meridian

Result is quite similar to reference:

When removing offset of 4 (including time correction) the 0 is not on top anymore:

When removing the 20 minute correction the x-axis has an offset:

Conclusion, we should keep the offsets as is, but the comments can be updated.