Missing "v"-component of wind from NAM GRIB files.

I commented on #45 a few weeks ago, but I decided to create a new ticket since that ticket is closed and most likely not being tracked. For reference, I was the original reporter of that Issue. Here is the comment from that ticket:

I’m just now revisiting this, and I’m still experiencing the same bug where the u winds are present but the v winds are not. This is using cfgrib version 0.9.7.3. Below is the output from opening a NAM awphys** GRIB2 file which can be found within this database. I’ve made sure to check all the xarray.Dataset objects output by the cfgrib.open_datasets method and none contain v winds.

<xarray.Dataset>
Dimensions:        (isobaricInhPa: 39, x: 614, y: 428)
Coordinates:
    time           datetime64[ns] 2019-10-30
    step           timedelta64[ns] 00:00:00
  * isobaricInhPa  (isobaricInhPa) int64 1000 975 950 925 900 ... 125 100 75 50
    latitude       (y, x) float64 12.19 12.22 12.25 12.28 ... 57.39 57.36 57.33
    longitude      (y, x) float64 226.5 226.6 226.8 226.9 ... 310.3 310.4 310.6
    valid_time     datetime64[ns] 2019-10-30
Dimensions without coordinates: x, y
Data variables:
    t              (isobaricInhPa, y, x) float32 ...
    u              (isobaricInhPa, y, x) float32 ...
    w              (isobaricInhPa, y, x) float32 ...
    gh             (isobaricInhPa, y, x) float32 ...
    r              (isobaricInhPa, y, x) float32 ...
    tke            (isobaricInhPa, y, x) float32 ...
Attributes:
    GRIB_edition:            2
    GRIB_centre:             kwbc
    GRIB_centreDescription:  US National Weather Service - NCEP
    GRIB_subCentre:          0
    Conventions:             CF-1.7
    institution:             US National Weather Service - NCEP 

Additional information requested from your contributing guidelines:

• OS: macOS High Sierra v10.13.6
• Installed using pip with xarray==0.14.1

[xebadir]

Testing for GFS and RAP also confirms this is now resolved in 0.9.8.2. Great work on solving this.

[bbonenfant]

I just tested on a NAM grib and it appears that the u and v components are now available. Thank you for the continued support of this issue. I hope this turns out to be stable and reliable!