sfence

I have created this PR to fix #14110. If it is no longer interesting, please close #14110 as won't add. This PR has performance benefits in comparison with `core.get_node` in...

From the API side view, it can be nice to have a complement function to the `core.bulk_set_node` function. But at all performance benefits of #14384 are much better.

> could this be used for waterlogged nodes ? , if so is it possible to have a filter for where to cover a node (so we could waterlog only...

@Desour I added some info to `lua_doc.md`, but screenshots are probably still better to show what it does than a thousand words. If you have any additional questions, feel free...

> Ah, I see now. So for sunken nodes, you need to register a new node for each liquid x inner node pair. And this doesn't work if the inner...

New ideas in my mind about using param2 and field inner_node: It can be determined by `leveled_max` value, how many bites can be used for the inner node. And it...

The new version removed the need for node pair registration for visual purposes. Inner nodes are now defined by metadata fields `inner_node` and `inner_param2`. These fields are cached, so rendering...

@SmallJoker I do not think, that the `submerge` field will be enough to work in situations like this:  

> What version of devtest is this? This is an unmerged branch without the core developer's support. To try it, you must clone this branch and compile Minetest. But at...

There is a code https://github.com/minetest/minetest/blob/92c55c27cf520bd7b0e59639cc7cce7bcaf5d747/src/server/player_sao.cpp#L462 for players and similar https://github.com/minetest/minetest/blob/92c55c27cf520bd7b0e59639cc7cce7bcaf5d747/src/server/luaentity_sao.cpp#L329 for Lua entities. So, is this only a documentation issue, or what is expected from punch methods to do, if...