> Now people are interested in my research and inviting me to come back and do a PhD, but I’m not sure I can ever go back to having to wake up early in the morning, sit in a boring seminar, be pressured to take on teaching duties, and not be able to escape to a warm country in the cold winter.
I understand that this may be specific to your sub-field, but this is absolutely not the case for many students. It varies heavily with university and country. I believe it makes a big difference whether you're considered to be a staff member or a student. In the UK you're a student, in Germany, you're staff.
We have several part-time PhD and Masters students in our department and I think the requirements are generally relaxed for them. Most of them work remotely. Many of them also have other jobs, families, etc. One is a sci-fi author!
Here, teaching is something that students occasionally do to earn a bit of money. It's rare though, usually it's only lab supervision or grading - we pay lecturers to teach. So this is absolutely not the case for a lot PhD positions. In the US and other European countries this is different and requirements will vary.
Our lab's core hours are 10 AM - 4 PM. When you go into work is mostly dependent on your supervisor. In the majority of cases they're laid back and as long as research gets done, they're happy. I routinely arrived at midday and left at 6-7PM during my PhD (and I do the same in my postdoc). Similarly nobody cares if you need to take the morning or afternoon off for personal reasons. That said, I know some people who have more traditional supervisors who treat their group like a sweat shop.
Students funded by research councils here get around 8 weeks of holiday a year. I don't think anyone counts though and it's rare for anyone to actually book that much time off. As a postdoc, I get 30 days, plus up to 5 carried over (and that doesn't include forced closures over Christmas). Staff can accrue up to 35 I think. Winter sun is definitely an option. I knew someone at Oxford (mathematics) who took 3 weeks off and just shrugged when we asked if his boss knew where he was. I confess to taking the odd week off, forgetting to tell anyone, and having to call into an unexpected telecon from my campsite.
We do make PhD students attend seminars, but that's mostly to get them involved in research, asking questions, etc. Boring is I guess a matter of opinion - you're expected to hear what other people in your field are doing. On the plus side, our seminars are either on Wednesdays, followed by subsidised dinner with the speaker; or they're on Fridays and followed by pub in the evening.
> I know that in hard sciences, you cannot meaningfully contribute to the field without access to a lab and other such infrastructure, and therefore you have to go back to a university position.
Well this is definitely not true, unless you deal with instrumentation or experimental work. You need infrastructure for sure, but unless you're actually doing practical work that can be remote access. Most students here log into our cluster to do work.
> I talk to my colleagues and get the impression that I have more time for research than they do, because anyone in a formal position is saddled with bullshit administration duties.
This bit is unfortunately true. The higher up you get, the more admin you need to do. And once you go permanent, teaching does become a necessity in most departments.