At scale, if the earth was the size of a basketball and the moon the size of a baseball, they would be about 7.5m apart.
For these images, the moon is not directly between the sun and the earth, so it cannot block sunlight from hitting the earth. So, no shadow.
This image might better help explain. Since the earth's orbital plane is in a different xy plane than the moon's, the moon crossing over the reflection guarantees nothing about whether it crossed the sun's emitted light.
http://www.cnyo.org/wp-content/uploads/2014/04/2014april10_e...
As some other folks have mentioned, the Moon's orbit is tilted a total of 5.2 degrees[2] relative to the Earth-Sun plane, so DSCOVR will see quite a number of DSCOVR-Moon-Earth alignments, but never a Sun-DSCOVR-Moon-Earth alignment.
[1] http://www.nesdis.noaa.gov/DSCOVR/
[2] http://hyperphysics.phy-astr.gsu.edu/hbase/solar/solecl.html
So, even though a perfect alignment of DSCOVR (at L1), the moon, and Earth would mean a shadow and thus eclipse, the moon must be a little off that alignment on this pass, throwing its shadow into dark space, even though it looks very nearly aligned in the photos.
The article says : "Combining three images taken about 30 seconds apart as the moon moves produces a slight but noticeable camera artifact on the right side of the moon. Because the moon has moved in relation to the Earth between the time the first (red) and last (green) exposures were made, a thin green offset appears on the right side of the moon when the three exposures are combined. This natural lunar movement also produces a slight red and blue offset on the left side of the moon in these unaltered images."