There are no sports watches with enough HRV accuracy for all day in a minute-by-minute resolution. During movements PPG signal is distorted and RR intervals cannot be precisely calculated. However count of RR intervals are less impacted by these distortions and HR from PPG during major movements can have acceptable accuracy.
Oura ring measures HRV during stationary periods, when major movements occurs it cannot accurately measure HR and HRV. There are no PPG or RR intervals data, meaning that its processed in a black-box style.
Polar H10 is not comfortable for 24/7, because of tension of chest strap. I slept 2 nights with it and its really annoying.
Some devices may have short term HRV snapshot, but i would recommend using Polar H10 for that. I have 3 years of morning 1 minute standing HRV measurements from Polar H10 and Movesense MD and these devices are reliable enough for this kind of test.
Garmin, Apple Watch, Samsung etc does not provide all day HRV data, due to limitations of PPG signal during movements. I’ve tried different PPG devices with raw data access (Polar Verity Sense, Shimmer GSR+, Hypnodyne ZMax) and i dont see how postprocessing can overcome major noise in signal, which results segments full of noise and there are no way to detect peaks when you only see noise. At least at current PPG sensor generation, maybe it will improve in a future. But if you are good with only having all day HR and some idea about HRV during sleep period - i would recommend getting Fitbit Charge 6.
I’ve build full 24/7 framework for measuring ECG/HRV with raw data collection and fully automated processing. You can read about that in my blog post ECG & HRV 24/7 with some science. I dont see more casual way of getting all day HRV with acceptable quality which covers full day without major gaps.