Better to say the vast majority of them are later viral shedding or very minor, probably not contagious cases.
There's probably been a few reinfections, and some decent documentation of a couple. If you repeat an experiment millions of times on diverse individuals, you can expect some variation.
There have been cases when genomic analysis was performed during the two infections and it "showed genetically significant differences between each variant associated with each instance of infection".
So reinfection actually is possible.
https://www.thelancet.com/journals/laninf/article/PIIS1473-3...
https://www.nature.com/articles/s41467-020-19802-w
"There was a low repositive rate in recovered COVID-19 patients in Wuhan. Results of virus culturing and contract tracing found no evidence that repositive cases in recovered COVID-19 patients were infectious, which is consistent with evidence from other sources."
Same paper also states that asymptomatic cases are much less likely to be infectious.