Make skin and bone compatible with attach

[jjyyxx]

The feature, motivation and pitch

deformable/skin/bone uses body field to set Name of the body corresponding to this bone., but when attaching a skinned body hierarchy to a parent model, the attach prefix changes the body names, while the skin still looks for the original names, resulting in unknown body xxx in skin. It could be improved to have a more coherent attach API.

Alternatives

No response

Additional context

No response

[darknight0847]

I have tried to solve the issue but the engine files are not giving errors, is there anything I am missing?

[physical-graphics]

I've also ended up interested in this issue, after first having similar issue with attach and flexcomp (#2977)

Today, when adding more detailed skins to the soft tyres of my simulation-in-progress (https://youtu.be/wiThVeOMFWg?si=aPhufmuUKl5OV0-o) I encountered this problem of propagating prefixes to skin and bone elements.

After adding a bit of printf mess here and there, inspired by @quagla's comment in review of @darknight0847's PR #2937, I got console readout that showed:

• mjCSkin::NameSpace wasn't called at all during successful load of model with 4 flexcomps and 4 skins (I manually made 4 .skn files with prefixes baked in to bone names to get past this issue temporarily)
• mjCFlex::NameSpace was called during this load, once or maybe twice per flexcomp
• the CopyFromSpec member functions get called a lot of times, for both mjCFlex and mjCSkin
• for most of those calls, the spec_item_s that I tried monitoring were empty, for both mjCFlex and mjCSkin
• for mjCFlex, the calls of mjCFlex::CopyFromSpec that came directly after mjCFlex::NameSpace had the correct number of items
• for some reason I can't get any results in a call hierarchy check for calls to mjCFlex::NameSpace, which I know is being called, to try to get an idea where to look for missing calls to mjCSkin::NameSpace

Does this move consideration of this issue forward at all?
Any idea of where the missing call to mjCSkin::NameSpace might be supposed to come from?