grc: snippets before init is called

[mormj]

Description

There are situations where snippets need to be inserted before block initialization, such as when
a block constructor makes contact with an external server, it might be handy to initialize the server
or check whether it is ready prior to calling the flowgraph init function that calls the block constructor

The problem with the current construct is snippets can only be injected after the flowgraph is initialized,
since they use self as a reference to the flowgraph object. There is probably a better way to do this
but the easy solution is to pass None as what will become self, and snippets that are inserted before
main are not able to make calls to self

Notes:

1. Main Before Init is placed second on the list, so Main After Init is the default

I'm wondering if it would be more useful to have the injection point inside the top_block.__init__ after variables
are instantiated, then the snippet would have access to variables

Related Issue

None

Which blocks/areas does this affect?

Just GRC when snippets are used

Testing Done

Checklist

• I have read the CONTRIBUTING document.
• I have squashed my commits to have one significant change per commit.
• I have signed my commits before making this PR
• My code follows the code style of this project. See GREP1.md.
• I have updated the documentation where necessary.
• I have added tests to cover my changes, and all previous tests pass.

[mormj]

Perhaps just doing the second commit here is cleaner, which looks like:

        ##################################################
        # Blocks
        ##################################################
        snippets_init_before_blocks(self)
        self.blocks_throttle_0 = blocks.throttle(gr.sizeof_gr_complex*1, samp_rate,True)
        self.blocks_null_source_0 = blocks.null_source(gr.sizeof_gr_complex*1)
        self.blocks_null_sink_0 = blocks.null_sink(gr.sizeof_gr_complex*1)

[willcode]

There's no harm in adding more snippet hooks. There's got to be a better way to insert more complicated user code, though.

In the short term, we could have a block where the user can fill in more than one hook, so things stay more coherent, and keep compatibility.

[mormj]

In the short term I think the second commit is safer than the first, and maybe also should have one "init before connections". But you're right there's got to be a better way to bring it together