With the scoring rules as written, I would tag it with code-challenge only.
From the code challenge tag wiki (emphasis mine):
The most common problem specific to a code challenge is that its winning criterion is highly subjective (i.e. each answer is equally valid) or that multiple, conflicting winning criteria exist.
In addition, the code golf tag wiki excerpt says:
A competition which scores on a mixture of source length and some other criterion should be tagged [code-challenge] instead.
You can see several examples of this sort of criteria by browsing the code challenge tag. For example:
What's my name? Produce the name of the language indirectly
When does (x == x+2)?
Implement ROT-47... in ROT-47
Convert number to comma separated number
Arbitrary Base Conversion
All have some degree of subjectivity in addition to the other criteria. In some cases, votes are simply used as tie-breakers. In others, there seem to be no criteria at all, and the tag was edited in by someone later because it lacked one.
To be honest, code-challenge could be considered a bit of a mess, but it's useful to have something to fall back on when none of the other tags fit quite right.