allow referencing previous c-vars when resolving

[snopoke]

This allows you to split up complex logic in cvars into multiple steps and avoid repeating logic.

Here's an example where the 2nd var references the value of the 1st var.

<c-vars
    :slot_is_empty="{{ slot|length|yesno:'false,true' }}"
    :make_square="{{ square|default_if_none:slot_is_empty }}" 
>/

[wrabit]

@snopoke I was initially skeptical about allowing this because of the potential additional load this could introduce. For example, some component's I create have ~10 cvars, this component may be placed on the page 10 - 20 times, so that equates to around 100 / 200 additional instances of manipulating context vs the current state. But, after performance tests, I am seeing this change brings a render speed increase of ~30% (!). So please bear with me whilst I do some head scratching and looking into just why that could be and ensuring the benchmarks are up to scratch.

[snopoke]

@snopoke I was initially skeptical about allowing this because of the potential additional load this could introduce. For example, some component's I create have ~10 cvars, this component may be placed on the page 10 - 20 times, so that equates to around 100 / 200 additional instances of manipulating context vs the current state. But, after performance tests, I am seeing this change brings a render speed increase of ~30% (!). So please bear with me whilst I do some head scratching and looking into just why that could be and ensuring the benchmarks are up to scratch.

Thanks for the note. 30% increase is fairly significant so definitely worth looking into. Let me know if I can assist.

This change isn't essential since there are other ways to accomplish it, it's mostly for convenience. If we don't end up merging this I think it would be useful to add a note to the docs to make it clearer that this doesn't work.