Remove redundant 'reduce' overloads

[ghost]

Fixes #25454

[mhegazy]

Well.. with the removed overload there is no contextual type for the initial value parameter...

@rbuckton, @DanielRosenwasser and @weswigham thoughts?

[sandersn]

The contextual type would be incorrect anyway in the case that U !== T, wouldn't it? And I don't think initialValue will commonly be context sensitive. Except in the case of tuples, maybe.

Edit: Shouldn't inference be able to pick up U from the return type of the reducer function and use it as the contextual type? I don't think we're losing much here. Maybe running on the user tests is a good idea though.

[ghost]

I did come up with a test case that's broken by this, see the bottom of arrayReduce.ts. Probably not a common situation though... I don't think there's a way we can support initialValue having an arbitrary type, and also give it a contextual type based on the assumption that it's the same type as the array elements (which may be a supertype of what you wanted initialValue to be, which caused the original issue).

[ghost]

@mhegazy @sandersn @rbuckton @DanielRosenwasser @weswigham Could you review?

[ghost]

@mhegazy @sandersn @rbuckton @DanielRosenwasser @weswigham Could you review?

[mhegazy]

I am still concerned about the valid patterns that will be errors with this change.

[RyanCavanaugh]

@Andy-MS let's get this up to date and then see what the RWC run turns up

[RyanCavanaugh]

@typescript-bot test this

[typescript-bot]

Heya @RyanCavanaugh, I've started to run the extended test suite on this PR at ab17070. You can monitor the build here. It should now contribute to this PR's status checks.

[RyanCavanaugh]

Nontrivial RWC breaks occurred

[typescript-bot]

The TypeScript team hasn't accepted the linked issue #25454. If you can get it accepted, this PR will have a better chance of being reviewed.