This is just a wild guess, but I believe implementation still needs 3 month, so I would say 2020Q3. Features that are deemed to be stable enough will be merged earlier into the master branch.
Yes, you should take a look at both Core4INET and nesting. Both projects attempted to cover some part of the requirements for TSN networks, however implementing the needed features is a bit more complicated and cannot be implemented properly on TOP of INET. Proper functionality requires that it should be implemented in INET. That's what is happening now, we are taking the most important features and trying to implement it inside INET (properly). There are 3 main thing needed:
- Queueing, traffic generation, shaping etc. This is done and ready in INET 4.2
- Clock modeling (planned for INET 4.3)
- Frame preemption (planned for INET 4.3)
In short, if you are in a hurry, take a look at both projects and pick the one that suits your needs. If you plan something that has a longer lifespan, you may be better off to base your project on INET 4.3 once it's available. (however note that INET 4.3 will conain only the basic infrastructure, there wont be any actual protocol implementation present). We hope that once the infrastructure is in place, the community will come up with the necessary models.