I'm not an electrical engineer, but wiring something other than ground to the ground pin is...well, I suspect it's straight-up malicious, although why you'd want to do something like this, I don't know.
Hanlon's razor: "never assume malice when stupidity will suffice."
It doesn't even make rational sense for a business creating cables to intentionally damage equipment. They'd just get sued, tons of bad PR/reviews, and gain nothing obvious.
More likely they just had production line issues, lack of QA, and poor training for staff. Resulted in malfunctioning equipment being sold. Most people that build cables don't understand how the cable works, they just follow instructions they're given (e.g. "red cable into position #1, white/black cable into #2, yellow into #3," etc).
Ultimately this might just be one incorrectly assembled cable; but the issue here is that they don't QA cables before they leave the shop. Electrical testing on most cables is quick, inexpensive, and automatable. They likely saved a few cent per cable by skipping it, but in the cable industry that might be significant savings.
I still wonder, what's in the minds of people who sell this kind of crap, and where did their conscience go? I wish we'd have a reliable way of getting rid of such vendors.
If engineer man does that, then he is not a good guy.
If engineer man does that, then he is a scummy asshole.
Secondly, the way USB declares that something is a host is by tying ground to the sheath.
So, there's precedent for tying things together (through a passive), and tying things to ground (but only things that are floating). MY guess is that they bought some connectors designed for USB host that had the sheath tied to ground and in their attempts to tie D+ and D- together shorted to ground. Only a guess though, you really don't want to do this.