sql: require super user return status

[Gurio]

Resolves #32662 as discussed with @knz (hopefully correctly)
First commit is taken from #38124 which helped to find all the necessary places where code should be changed
I didn’t check if it builds with the last fix-commit, so please don’t review until TeamCity succeeds.

[CLAassistant]

All committers have signed the CLA.

[cockroach-teamcity]

This change is 

[knz]

Thanks the end result looks about right.

You're missing a piece though; as-is the code does not compile.

Then after that ensure the PR contains just your commit (instead of appending to the previous author's commit).
I could also recommend squashing the commits into one, there is no need for the intermediate situation which modifies many files to change the return values of RequireSuperUser.

Reviewed 3 of 20 files at r2, 1 of 1 files at r3.
Reviewable status: complete! 0 of 0 LGTMs obtained

[Gurio]

@knz I've updated it and run test locally, should be fine now, lets see what CI says.
two open things:

1. I've included @mgm702 's commit, to give kudos to his work. But as there is a problem with CLA, i will probably squash it into mine.
2. I've updated RFCS/20171220_sql_role_based_access_control.md, according to the current AuthorizationAccessor interface. But I am really unsure if it is the right thing to do, because old RFCs probably should stay as they were. What do you think?

[Gurio]

Ok build still fails, I’ll take a look

[knz]

RFC change is OK.

You can mention the name of the other contributor in the commit message, no need to include the commit itself.

[Gurio]

@knz: I think PR is ready. Could you check/merge?

[knz]

Reviewed 3 of 4 files at r4.
Reviewable status: complete! 0 of 0 LGTMs obtained

[knz]

Thank you for your contribution!

bors r+

[craig]

Build succeeded

• GitHub CI (Cockroach)