Thanks for your inputs even though some of those are contradictory
No, there are no contradictions at all.
Instead we just have you repeatedly ignoring time zones.
The sheer number of times you ignore it shows that it isn't a case of you just making a mistake.
Globe calculation gives 6° - 0.75° = 5.25° AKA 21 minutes of sunray travel after 06:00.
No, it doesn't.
That only works if you assume the sun should rise at exactly 6 am on an exact multiple of 15 degrees.
The only time such a calculation would be meaningful is comparing 2 locations with the same latitude and altitude. Not comparing it to an arbitrary point you say should be at 6 am.
It also requires the equinox to last an entire day, rather than being a single point in time. Better comparisons would be at the solstice, when the change in day length is shortest, not longest.
At the equinox, in Hong Kong, the length of the day was changing by over 1 minute each day. So that is an extra ~40 seconds of uncertainty to add to any values based upon sunrise and sunset times.
Again, due to Earth's orbit not being a perfect circle, and due to Earth's axis being tilted, the length of a solar day will not be 24 hours. Instead it will vary throughout the year, with the solar time sometimes being ahead or behind a "standardised" solar time where each day is taken as 24 hours.
If you want to have a meaningful comparison between different latitudes or altitudes, you need to focus on solar noon, not sunrise.
This is because sunrise is also affected by latitude and altitude. If you are on the top of a tall mountain, sunrise will be earlier and sunset will be later.
If you are at a more northern latitude, then as long as it isn't winter (or on the winter side of the equinox), then sunrise will be earlier and sunset will be later.
So if we make an honest comparison, we have London, at roughly 0 degrees east, with solar noon at 1153 UTC, being 8 arcminutes west of the "#true UTC0".
Now Hong Kong, being 114 degrees and 11 arcminutes east, puts it 5 degrees and 49 arc minutes west of "#true UTC+8", or an extra 5 degrees and 41 arc minutes.
Those 5 degrees account for 20 minutes, and with each degree being 4 minutes, each arcminute is 4 seconds, so those 41 arc minutes mean 164 seconds or roughly 3 minutes, bringing our total to roughly 23 minutes.
Also note that these values are reported to the nearest minute so you have plus or minus half a minute for each value. And with it only being 164 seconds rather than 180, it should be on the lower side.
So that means solar noon in Hong Kong should be around 1153+23 minutes UTC8, i.e. within a minute or so of 1216 UTC+8.
Now what is it? 12:15.
So well within the uncertainty of the reported values.
The day before the equinox UK still has solar noon at 1153 UTC0, and Hong Kong has it at 1216 UTC+8
So in quite good agreement with the globe.
In fact, just picking any random time, like today, we have London having solar noon at 1205 UTC+0, and thus Hong Kong should be around 1228 UTC+8, and what is it? 12:27.
Now we can also do a rough approximation based upon sunrise and sunset times, but that will be off due to sunrise and sunset depending on the latitude.
Again, on the equinox, we have sunrise in London at 0547 UTC0.
In Hong Kong, we have it at 0612 UTC+8.
This is a difference of 25 minutes. Not bad when you have the expected 23 minutes for difference in longitude from the "reference", and London being at a more northern latitude, and the uncertainty of roughly a minute from the rounding of the values, and the uncertainty of a minute from the change in day length in Hong Kong, and the uncertainty of 2 minutes for the change in day length for London.
Doing the same for sunset we have 1757 UTC0 for london and 1819 for Hong Kong. This time, 22 minutes.
Just like one might expect, the difference is larger for sunrise than sunset. As sunrise should occur earlier and sunset should occur later for the more nothern London.
And as the equinox is at 9:30 pm in Hong Kong, the difference between actual and "reference" time differences should be larger for the morning. But with all the uncertainty, we can't tell which is actually the bigger difference.