We absolutely should look to traditional ag/etc. to help provide wisdom and trends in the cannabis genetics market. My background is in biology/chemistry with a focus on plants, so the trends of agriculture market development are infinitely fascinating to me.
The reason breeders don't release F2s is two fold. First they don't want people getting plants that are very close to pure copies of the P1 generation, as you stated. The second reason is that most customers are very upset when they grow an F2 and the end result is nothing like the advertised hybrid (<-This is their bigger concern, since no one likes angry customers). The fact that so many "breeders" are releasing F1 crosses that can quickly be turned into F2s shows that they aren't as concerned with protecting their "asset plants" from duplication as they use to be.
I've lost track of how many "breeders" I've met at events, ranted loudly for 15 minutes about how being a flavor-of-the-day pollen chucker only produces multi-polyhyrbid garbage that is literally ruining the cannabis species, and then looked at their seed lists to see they are as guilty as it gets. Sadly, these pollen chucking get-rich-quick charlatans are more prevalent than the breeders of stable genetics these days.
To add supporting evidence to @StrainVenator's theory, in just the last 2 years there has been a massive increase in the number of reliable online sources for seeds. Many of them newly available to the U.S. market with no need to risk customs inspections. In addition to this, the types of strain crosses are mind bogglingly large, reducing the previous need of home growers to make new crosses to obtain the desired variety qualities. A second point is that larger and larger portions of the U.S. market are produced less by unregulated markets, which tend to sell seeded bud at a decent pace, and more by regulated commercial producers, which have a huge market demand for bud with no seeds at all. Seeded bud still happens in these commercial grows, but all those seeds end up in the trash at the concentrate extraction facility. I believe the legalization of cannabis across the U.S. is doing more to kill bag seed availability than the reactions of growers to the plethora of seeds available from seed banks is.
Being in a legal state and a zealot for trying new and old strains, I've realized that every flavor new breeders are "creating" has already been done with older strains in the last 40 years. In fact, many of those older strain flavors need to be recreated, because the originals aren't available anymore. This might be the one good side effect of having so many crosses available in seed form.
I agree, the requirement for homozygous genes is only for the specific traits being bred for/patented and having seeds that breed true 90% of the time is enough usually; meaning that each plant is likely only homozygous for the few traits selected and nothing else, resulting in a large percentage of heterozygosity across the entire gene code.
It's my understanding that for cannabis the big trait they need to lock down is terpene/cannabinoid profiles, which are all over the place for most current strains. Most of these current strains fail miserably at even the "uniform" or "stability" factors when viewed under the stringent scientific requirements demanded by traditional scientists.
The Blueberry Blast hazy smoke is treating me right tonight!