Super-linter: Add more pre-release testing

[bazzadp]

Is your feature request related to a problem? Please describe.
The GitHub super-linter has published a number of bad releases since go-live which have been left in place for hours and days. This is particularly a problem as this project does not follow versioning (see #708 ) so this affects even those repos that do not upgrade, and also affects all forks and branches meaning even if we try to work around a bad release it requires all those forks and branches to pick up the latest workaround. This is a big problem when all linting workflows across many projects suddenly break.

The latest example is #938. While that issue can be dedicated to fixing that immediate problem I would like to raise this separate issue to track how this was able to get through to a release. Consider this almost the RCA and follow up actions to try to prevent #938 happening again.

Describe the solution you'd like
Add test cases to prevent the cause of issues like #938.

Describe alternatives you've considered
The super-linter should also:

• Fix versioning - #708
• Add post release checks to immediately identify issues on a bad release - #948

Additional context
Issues that have broken the linter (not an exhaustive list):

• #704
• #938

[admiralAwkbar]

@bazzadp i agree... I take full responsibility for some of these releases and some of the lack of testing.
As this is more widely adopted, we need to make sure master/main is always hyper stable.
I will look into this week to see if we can create some additional safeguards to help resolve this

[ferrarimarco]

1. One idea here might be to fail with a fatal error if the TAP file is missing, instead of raising a warning (that no one is going to fix). Thoughts?
2. Set ERROR_ON_MISSING_EXEC_BIT to true

[admiralAwkbar]

@ferrarimarco I agree