Stride Length accuracy in FR 965 and general Garmin watches?

Stride length is a single-step length. I don't see a reason for it to be two-step length.

There are some research papers and websites said stride length is two-step length ? I was confused

https://www.healthline.com/health/stride-length

It's funny that several health and running sites are careful to distinguish step length and stride length, explaining that stride length is always the length of one revolution of a complete gait cycle (i.e two steps).

Then there's this:

https://marathonhandbook.com/average-stride-length/

A step is half of one stride, so there are two steps for every stride.

https://marathonhandbook.com/how-to-increase-cadence-while-running/

While more familiar to cyclists, the concept of cadence also applies to running. Your running cadence is your stride rate, or how many steps you are taking per minute as you run.

The same author wrote both articles.

Considering that cadence in running is always measured in steps per minute (afaik *), I'd say that step length is more useful than two-step length. Just too bad that the phrase "stride length" is ambiguous.

(* People always say to target 180 steps per minute, not 90 strides per minute.)

If that the case, stride length metric in Garmin watch is actually step length?  As my original post, 5km run, my steps recorded was 6666 steps. 

Yep. In fairness to Garmin, even running sites can't keep their terminology straight. In the example above, the same author said "a stride is two steps!" and "stride rate is steps per minute!"

Obviously nobody really agrees on a single definition of the word "stride" for running (at least in an informal context).

Exactly, step length is a better terminology than stride length. I read articles that Eliud Kipchoge stride length is 2m, almost 1.25 x his height, is this stride length or step length?  

That's his step length. If 2 m was his stride length (to use the strict definition of term), that would imply a step length of 1 m which is not very impressive. (My step length is 1 m when I'm out of shape and running slowly, and I'm far from being a pro.)

Another way to do a sanity check on the numbers is to multiply (assumed) cadence by step length.

If his step length was really 1 m and his average cadence was 180 steps per minute (180 is the baseline for a decent runner), then he'd be going 180 metres per minute which translates to 233 minutes (3 hours and 53 minutes) for a marathon. In order to run a marathon in his actual PB time - 2 hours 1 minute 9 seconds - his average cadence would have to be roughly 347 steps per minute which is humanly impossible.