VRADH and DBZH moments may be loaded from different sweeps, and DBZH-only sweeps dropped

[adokter]

Certain older NEXRAD data has different range resolution for lowest three (250 m) and highest two (500 m) elevation scans, as well as different resolutions for different moments, e.g.
https://noaa-nexrad-level2.s3.amazonaws.com/2010/09/11/KBGM/KBGM20100911_012056_V03.gz:

moment	range	azim	elev
R	1000	1	3.5, 4.5
R_HI	250	0.5	0.5, 0.5, 1.5, 1.5, 2.5, 2.5
V	250	1	3.5, 4.5
V_HI	250	0.5	0.5, 1.5, 2.5

• When multiple scans at same elevation are available, WSRLIB uses largest PRF sweep for radial velocity and spectrum with, and lowest PRF for reflectivity. I think a better approach would be to keep the reflectivity scan acquired simultaneously with reflectivity, so largest PRF sweep in all cases. However, currently the behaviour is unclear, because unclear which of two sweeps at identical elevation is returned by RSL_get_sweep(sweep, elev). Have to add control over this in vol2bird.
• Related: the above file includes multiple reflectivity scans that do not have associated radial velocity. These are currently not used or stored in converted ODIM output. Add access to these