Collections: better logging for chunking APIs

[josephjclark]

When we set(), we get logs like this:

ADA Collections: uploading batch of 1000 values to "nh-demo"...
ADA Collections: set 1000 values in "nh-demo"
ADA Collections: uploading batch of 1000 values to "nh-demo"...
ADA Collections: set 1000 values in "nh-demo"

And when we get(), logs are like:

ADA Collections: fetched chunk of 1000 values from "nh-demo"
ADA Collections: fetched chunk of 1000 values from "nh-demo"
ADA Collections: fetched chunk of 643 values from "nh-demo"

In both cases, we should find a way to neatly log the chunk and the total.

It looks a bit weird if we only get one chunk to do this:

ADA Collections: fetched chunk of 165 values from "nh-demo"
ADA Collections: fetched 165 values from "nh-demo"

So either we:

• Only log chunks if we need to (which is just a bit of fiddly code). This results in inconsistent logging
• console.debug the chunks and console.log the results. But a) Adaptor debug logs are weird kit#815 and b) lightning doesn't render debug any differnetly to info, so there's not much of a net gain for the user.

I could also not log the chunks, but its nice to see signs of life during big uploads.

Yet another option would be logs like

ADA Collections: fetched 1000 of 1643  values from "nh-demo"
ADA Collections: fetched 1643 of 1643 values from "nh-demo"

But maybe that's still kind weird