Fix lint errors across the repository

[gaganhr94]

Currently the lint is checked on the diff in a PR. There are several lint issues currently in the existing codebase which can be fixed in a phased manner.

102 issues:

• errcheck: 50
• staticcheck: 50
• unused: 2

[gangadhar-res]

Hi @gaganhr94, I'd like to pick this up. Since you suggested a phased approach, my plan is to start with the 2 unused findings plus a first batch of errcheck fixes (roughly 20-25, grouped by package) so each PR stays easy to review, then follow up with the remaining errcheck and staticcheck batches. If you'd rather I slice it by linter type or by package, happy to do that instead. I'll open the first PR in the next day or two.

[gaganhr94]

@gangadhar-res are you still interested to continue working on this issue ? I was planning to assign this issue to you if you are interested

[gangadhar-res]

yeah assign this to me, I will take a look sometime this week.

[gangadhar-res]

Hi @gaganhr94,

Before I dive in do things, I reran the linter on main (26c3eaf) with the output caps off,
since the defaults stop at 50 per linter and hide the real size.
It's 394 findings: 217 errcheck, 177 staticcheck. arangodb has 167 and neo4j 88,
so those two are about two thirds of it.

Two questions before I write more PRs.

neo4j already has unused and forcetypeassert excluded with the note that this backend does
it a lot. Some of its 88 look like the same story, so would you rather exclude those than
fix them?

On staticcheck, 116 of the 177 are three rules: SA1019 (45), ST1005 (36), ST1023 (35). The
last two are mechanical and I could do them repo wide in a PR or two. SA1019 I'd keep
separate and go slower, since swapping a deprecated API can change behaviour.

Next up I'd do pkg/assembler/backends/keyvalue, 19 findings in one package, same shape as
the last one. Then arangodb as its own series.