`measurementStopDate_1`, `time_1` and `time_difference` are wrong for S1 L2 OCN products

[vincelhx]

measurementStopDate_1 returns the start date and time_difference is affected (tens of seconds). Indeed the SAR acquisition dates are recomputed after passing start date in every cell and after resampling.

here

coloc_sat/coloc_sat/sar_meta.py

Lines 54 to 58 in d52a603

	ds["time"] = xr.DataArray(
	data=np.full(ds["lon"].shape, t),
	dims=("owiAzSize", "owiRaSize"),
	attrs={"long_name": "time", "standard_name": "time"},
	)

here

coloc_sat/coloc_sat/intersection.py

Line 900 in d52a603

# FIXME astype unknown effect on bool or else

and here

coloc_sat/coloc_sat/intersection.py

Line 1262 in d52a603

unique_time = np.unique(ds[meta.time_name])

unique_time = np.unique(ds[meta.time_name])
ds.attrs["measurementStartDate"] = str(np.nanmin(unique_time))
ds.attrs["measurementStopDate"]  = str(np.nanmax(unique_time))

• solution : keep time_1 as a constant, remove it from variables and ensure that time attributes are kept