Add a cache for decrypted values

[stlava]

Addresses #239

[rnelson0]

@stlava I've fixed the tests, if you can rebase this to make sure it passes.

I do like the idea in #239 of an option to control whether caching is enabled. Do you think you could add that to this PR?

[stlava]

@rnelson0 I added an option. It didn't seem very necessary to include this part of the sub-command options but let me know.

[rnelson0]

This looks great! I'm hoping @diranged can give us some feedback shortly so we can see how well this works. If there's no response within a week, I think we then merge it and revisit performance later if it's suffering. Thanks!

[stlava]

@rnelson0 I work with @diranged and took over the ticket. I deployed this yesterday at 22:00 utc. Here's the trend:

[rnelson0]

Oh, cool! That's noticeable, but not that drastic of a change. Is your compilation time coming from something else or does the caching need more work?

Happy to merge it, thanks for the contribution!

[stlava]

Adding the caching was a step forward for us to cut down the compilation time but as you pointed out the rest of our time does come from us hitting other external resources. Thanks for getting this in!