feat(script-results): build a map of results from the script executions against the repos list

[travi]

potential example of what #222 could lead to

[gr2m]

It's a different approach than I was thinking, but I like it. I thought to use logging as a way to create data and reports as a means to reduce that data to information. But using the return of a script is more universal 👍🏼

[travi]

so, the bit that isn't handled with this proposal is what to do with the results within the context of the cli. the results are returned from the programatic interface that can be used by other consumers, but i could use some help thinking through how to approach what is best to do in this consumer.

we are still setting the context on the logger if that function exists. should that change to instead coordinate logging outside the context of the programmatic interface? alternatively, we could continue setting the context as was done previously and defer adjustments to that until later? later could still mean before promoting the beta to stable, or it could mean further out

[gr2m]

I'll need to find a moment to think about this and right now I don't have much time. Is this blocking you or urgent in anyway? I'm super excited about moving this forward!

[travi]

Is this blocking you or urgent in anyway?

no, all good. i'm just thinking about a few different contexts where i think we could lean into this a bit more, so thinking through how to make these sorts of things generic enough to fit those other contexts. for now, more predictive of what is to come. when we do get to the point of acting on using directly, we would also be fine to use in beta state, so getting to that stage is valuable. don't feel like there is any pressure at all to get this to a stable release quickly

[travi]

i'm getting close to some situations where i need to inspect a bunch of repos for the current state of a few details without yet performing updates at scale. this change feels like it would fit that usecase well too. the inspection results could be returned as results of each execution of the script and the caller can then do what it wants with the returned results for the whole list.

if i could bring your attention back to this one when you have time, i'm curious what thoughts you might have that i havent considered. if it feels like a path forward, i'd love to understand how you feel about merging this to the beta branch so that it could be consumable in a published version. if it feels like the wrong path, i'm more than happy to think through other options together before that