Revamp verilogtree to include an interactive mode

[LiamSkirrow]

I've made the decision that verilogtree will become a larger project with more functionality. Therefore, it's important to include the functionality as standalone tools that can be invoked from the command line, as well as being able to use them in a more interactive mode with tab completing etc.

The standalone functions of verilogtree should still be useable via the command line, but now including the specific name of the command to be run:
verilogtree tree [args]
or
verilogtree trace [args]
etc...

Alteratively, running verilogtree as a standalone command verilogtree, will result in an interactive session being started, where you can load or reload a design, and when entering an instance's hierarchy, you can TAB complete to get the module names at that level:

$> verilogtree
[vt] > tree
top
├── mod0  
│   ├── mod3  
│   │   └── mod4  
│   └── mod1  
│       └── mod3  
│           └── mod4
[vt] > trace top/mod0 (user hits TAB)
mod3  mod1
[vt] > 

You get the idea, this would make for a really powerful tool, and would also be extendable for future tools that pop into my head,

[LiamSkirrow]

The use of the standalone commands should be identical to how things currently work with the hierarchy tree printing tool, just give the RTL source file path or a filelist, and that's it.

However, with interactive mode, if you try and run the tools straight away, it should give an error message saying that you haven't yet loaded the design. So you need to run the load_design command, which requires individual file paths or the path to a filelist. Once this is run, only then can the other commands be run.

The whole point of this, is such that when the design is loaded, it will actually parse the RTL at that stage so that when printing the tree or tracing a signal, it will happen very very quickly since the design is preloaded. When running the tools standalone, the parsing and processing of the design hierarchy and signals have to be done on-the-fly which is slow and laborious.

If the RTL is modified after it has already been loaded, then this necessitates the use of the reload_design command, which uses the same file paths or filelist already supplied by the user (whichever was given originally) to re-parse and process the design, probably from scratch.

[LiamSkirrow]

This needs to include the implementation of a split() function for strings, to this effect the cpp std::getline() function is probably good enough: https://en.cppreference.com/w/cpp/string/basic_string/getline

Just need to make sure that the source string only has a single space char since double-spaces (or more) would potentially be missed out by the split() function.