Hi Larisa,
In practice I tend to see papers where researchers either ignore that their timepoints are unequally spaced or interpolate so that they have equal spacing. Sometimes people put "NA" between the final observation of a day and the morning observation to ensure that lagged relations reflect only within-day changes.
Hope this helps. There's no definitive "right" way to accommodate this that I know about although I would love to hear from others on this.
Best,
Katie