First off, I will just say that USA's reliance on machines to cast and count votes is likely to lead to poor administration and abuse. In Canada, we use a paper ballot and a pencil. So do most other western democracies. The machine route has not proven to be more "honest."
But having said, a change to a paper ballot in the USA is not likely for some time.
Because this evidence did not find its way in front of judge, I have to assume that this evidence is not strong. So I say either this article is not telling the whole story or the legal team is inept. So I will just add some likely reasons why this particular case didn't go further.
1) Machines break down--even with the best and most honest operators. This machine just might have had an honest breakdown. You need proof that the poll worker deliberately altered the machine. Just because you say the machine can be altered does not mean that it happened.
2) And the finger points two ways. Would it not be possible for R-leaning poll worker to do the same in another county?
3) If Antrim County is the only county in Michigan with this kind of breakdown, this does not imply systemic fraud. Are there other machines with this problem? Were there other counties where the results were far off expected projections? Maybe if there were 10 such counties, Mr. Guilani might have found some success in the court room. But one case is not enough.
4) Where were the Republican poll watchers? While I don't like machine voting, I see big problems with a poll worker being able to recast broken ballots without poll watchers from both sides watching that process. If R watchers were chased out, then some iPhone videos would be pretty good evidence. If there were not enough R watchers, then put the blame on the Republican Party for not being able to find the volunteers.
5) The swing from Nov 3 machine count to Nov 5 hand count is significant. But so too is the swing in the Nov 21 count. Sixteen days past election day is too long, and I have to assume the possibly that--in this R-dominated county--some further tampering occurred. If I were a judge, I would go with the Nov 5 handcounts.
6) The Canadian/Chinese corporate connection is a moot point. The election officials (both R's and D's) decided to use these machines in 2020. I assume that they did some inspections beforehand and found them trustworthy. Trying to throw the results of these machines out would mean that the results of all other similar machines should also be tossed across the USA, including those in R-dominated electoral jurisdictions. In other words, going this route means there really is no winners anywhere: that includes congress, governors, and state legislatures, many of whom are Republican. Is that what you really want? Is the Trump presidency so important that you would put the legitamicy of all those other Republicans in doubt?
7) While I can envision that a poll worker is given a password to a voting machine to help with some troubleshooting, I find it unrealistic that such a worker would be given access to the depths of machine as to adjust the weighting--assuming that such an adjustment can actually be made. You would need a lot more evidence to prove this point. Maybe sequester one of those machines and show--in front of a judge--how easily it can be done.