Fix Vet Errors

[tchaudhry91]

What does this change

This PR is meant to eliminate the go vet warnings/errors in the codebase.
Present State:
1 big commit which fixes the most common recurring vet errors (unkeyed fields in composite literals).
There are 2 major culprits here:
ExtendedBundle which has an embedded Bundle struct. Relatively straightforward fix here to add Bundle before supplying the embedded struct. Even I didn't know this was a thing, always thought embedded structs had to be anonymous :)
Found the solution from here: https://stackoverflow.com/questions/49461354/go-vet-composite-literal-uses-unkeyed-fields-with-embedded-types

The other one is in the Primitive.E type in the mongo driver. We are generally passing Key and Value as a {"some_key", "some_value"} in all places. The recommended way is to go with {Key: "some_key", Value: "some_value"}. This feels a little extra verbose to me, but as highlighted in the link below, it offers good protection against changes in the library, so I suggest we keep it. Also fixes vet warnings :)
https://stackoverflow.com/questions/54548441/composite-literal-uses-unkeyed-fields

Update:
I have added another commit to resolve all the Struct Tag issues. Only minor typos here (although, it's surprising functionality wasn't broken in a few places). Nevertheless, all resolved now.

Update 2:
I have also renamed the example tests to use suffixes according to (https://pkg.go.dev/testing#hdr-Examples).
The old values were incompatible with Vet.

What issue does it fix

Makes incremental progress on #2081

Notes for the reviewer

This is a major chunk of the go vet warnings, but not all. I will resolve the rest via a separate PR. I kept this one segregated because of the theme.
Update: Also added a commit to resolve struct tag issues.

Checklist

• Did you write tests?
• Did you write documentation?
• Did you change porter.yaml or a storage document record? Update the corresponding schema file.
• If this is your first pull request, please add your name to the bottom of our Contributors list. Thank you for making Porter better! 🙇‍♀️

Reviewer Checklist

• Comment with /azp run test-porter-release if a magefile or build script was modified
• Comment with /azp run porter-integration if it's a non-trivial PR

[tchaudhry91]

I have added a comment regarding the broken unit-test. The malformed struct tag for yaml meant the file wasn't using the correct name. Will defer to someone with more knowledge to get the best way to fix this. Long term, we probably want to keep the correct tag and regenerate the golden comparison file. Not sure how much that will impact current consumers. Alternatively, we could use the "wrong" value in the tag if we want. Let me know.

[VinozzZ]

Thank you for submitting the PR to fix lint and vet errors!

I have added a comment regarding the broken unit-test. The malformed struct tag for yaml meant the file wasn't using the correct name. Will defer to someone with more knowledge to get the best way to fix this. Long term, we probably want to keep the correct tag and regenerate the golden comparison file. Not sure how much that will impact current consumers. Alternatively, we could use the "wrong" value in the tag if we want. Let me know.

Thank you for putting up the PR for fixing lint and vet errors. From my perspective,, we should fix all the schema issues while we are still in alpha. but we probably should wait for @carolynvs to confirm it

[tchaudhry91]

Hey all,
This PR is now complete from my point of view, here is a summary for the reviewers:

1. Fixes all the unkeyed fields in the composite literals.
2. Fixes the struct tags. This is a bit more complicated as this will technically be a breaking change? This is because the old tags were incorrect and hence ignored. So far, this has been detected in one place in a test file, but I assume others would also be impacted. For now, I have regenerated that test file and the tests pass now. If we want to go with another approach, let me know.
3. Renamed ExampleTests to use PackageName_suffix format as dictated by Vet.

The unit test failure in the PR seems to be due to a connection reset, i.e flaky. Can someone re-run it? I have confirmed that unit tests run fine on multiple machines.

[carolynvs]

Thank you for going over the codebase and getting all the vet errors fixed! Let me know if you'd like to add that constructor function and I'll hold off on merging.

[carolynvs]

We can always switch to a shorter constructor function, e.g. cnab.NewBundle(b).

I think we use extended bundles enough that this would be worthwhile to change. If you are up for it, it would be great to have that in this PR, but if you are too busy, we can do it in a separate PR.

[tchaudhry91]

Sounds good! I'll make this change.

[carolynvs]

Since we don't control the bson package, I don't think making a constructor function is necessary and using explicit keys when initializing E isn't too big of a deal.

Thanks for tracking this down!

[carolynvs]

This is why I'm super excited to have vet run by default soon. It will save me from myself 😝

[carolynvs]

I restarted the test run, it was a temporary glitch in the matrix. 😀

[tchaudhry91]

@carolynvs I'll add the constructor in this PR only. Let's hold off on merging till then. Shouldn't be long

[tchaudhry91]

@carolynvs I have added the constructor and made the replacements in all the non-zero initializations.
For consistency it's probably better to come up with a something to replace all the ExtendedBundle{} symbols too. Although, these are almost exclusively used to return zero values along with an error, in which case this value should never be referenced, however, for readability, I suppose a mix of NewBundle and ExtendedBundle{} might be a bit confusing. Nevertheless, if we want to replace those too, I can do in a new PR. This one is getting a bit unwieldy already :)

[carolynvs]

This looks great, thanks!

I don't have a problem with using cnab.NewBundle when wrapping and cnab.ExtendedBundle{} when we want an empty bundle. 👍