Contradictory type checks in Analyzer

[krisds]

As requested, I will split #38033 into separate PRs. Here is the third part.

This one resolves:

• Contradictory type checks

[elasticmachine]

Pinging @elastic/es-search

[costin]

Thanks for raising the PR.
However I don't think the proposed solution is correct; I can see how an analysis tool will pick it up (Literal and NamedExpression used to be different hierarchies which got unified at some point though they might be split again).
As the literal already inherits the NamedExpression behavior , the correct solution would be to remove the literal check all together - let me know if you want to pursue this or not.
Thanks,

[krisds]

Thanks. I was unsure what the right solution was. If you like I can remove the literal check and update the PR.

[costin]

Looks like I've missed the PRs and added here the comment that was made for
#38150.

This is another instance of code left behind from a structural refactoring (making functions named expressions). I'm on the fence on removing this since the code has meaning, especially as functions will be made again non-named expressions in which case the order is important...

[krisds]

So, should I remove the check, or will you leave it as is and close the PR ?

[javanna]

@costin could you please merge this or close it if you think the code is fine as-is? Thanks!

[costin]

Apologies for the late reply - this got lost in my inbox.
I'm opting for leaving this as is since it plays into an upcoming refactoring.
Thanks for raising the PR @krisds.

Cheers,