Many mammals on land seem to be quite dangerous to humans: e.g., tigers, lions, elephants, hypos etc. However, their sea-dwelling counterparts (e.g., orcas, dolphins, sea lions, whales etc.) are relatively friendly to humans (and in contrast to say sharks). Is there a simple reason for this?
More encounters on land
There are more land animals that meet humans walking around than there are sea animals that meet humans swimming around. If fewer attacks happen on humans in the sea environment, it might just be because there are fewer encounters with marine mammals than with land mammals.
Greater species diversity on land
In your list used as example, you cherry-picked a few aggressive(-ish) land mammals but the vast majority of land mammals are not aggressive toward humans. The fact that there are no marine mammal that are aggressive toward humans might just be because there are too few fewer marine mammal species.
Pseudoreplication and phylogenetic signal
I would argue that your sample size is not as large as you might think. There are only four lineages of mammals that evolve to live in the seas. There is therefore a phylogenetic signal and ignoring it would be a problem of pseudo replication.
For example, I could very well think of territorial behaviour (which is where a lot of aggressive behaviour comes from) as having a strong phylogenetic signal.
Why considering only humans?
Your post title talks about aggressiveness in general and the content of your post talks about attacks against humans. It sounds that considering only attacks against humans to be very unfair. Orcas, typically, are fierce and violent predators although no orcas have ever attacked a human in the wild.
Are they really that friendly to humans?
There has been a number of attacks in captivity (well known cases with orcas). Also, there have been a number of whales attacks against boats (according to this BBC article).
In short, I think your comparison is unfair. I think you are looking for an explanation for a pattern that does not really exist.