Dynamic kernel procedure invocation

[Fumuran]

This PR reworks the way syscall procedures being executed: instead of using syscall for each procedure, now we load the offset of the desired procedure to the stack and then execute syscall.exec_kernel_proc. More details can be found in the corresponding issue: #701.

TODO:

• Update docs
• Update CHANGELOG

[Fumuran]

I am putting forward a proposal to put on hold this task until the Miden VM 0.10 is released.

It is extremely complicated to make this feature work using the "pre-refactoring" version of the assembler. To make it work it's needed to put the hash of the desired procedure to the CodeBlockTable, which could be done by using one of the call, syscall or procref operations. call and procref works similar and can't be used with kernel procedures. The only way to achieve it is to call this procedure with syscall somewhere at the preparation stage and then roll back the stack, but considering that there are 29 procedures in the core overall, it will be a mess. But even if I manage to implement it it still will be needed to rework the implementation to make it compatible with a new version of the VM and assembly.

The other options is to try to change the dependencies of the miden-base so it will work with an updated version of the assembler. This is quite a complicated task, since a lot of breaking changes were made in the new version of the VM. But this is not the only problem: new assembly will require at least a new version of the crypro repo, so this refactoring will be pretty complex. And considering that current version of the VM is not even final for the 0.10 release, I think that it is not appropriate to spend time and effort for this refactoring.

Currently the core functionality for that feature is already implemented, so it shouldn't take too long to finish it after new version of the VM is released.

@bobbinth what do you think?

[bobbinth]

Thank you! Looks good! Not a full review yet, but I left some comments inline.

Also, I'm curious how this affects cycle counts for our transaction benchmarks.

[bobbinth]

Looks good! Thank you! I've added some more comments inline.

[bobbinth]

Looks good! Thank you! I left some more comments inline. The main one is about checking kernel procedure bounds.

[bobbinth]

Looks good! Thank you! I left a few small comments inline. Once these are addressed, we can merge.

[bobbinth]

All looks good! Thank you!