Make constructors for basic number types inferable

[martinholters]

Starting from the assumption that a constructor should return an instance of the correct type, i.e. T(...)::T where T isa Type, I wondered: Does that hold for all types in julia proper? (Answer: no, e.g. ref. #37186 for an accidental violation I stumbled upon.) And can inference figure it out?

This PR makes the answer to the second question "yes" for the basic number types, see the test:

julia/test/numbers.jl

Lines 2696 to 2700 in dca67c9

	@testset "constructor inferability for $T" for T in [AbstractFloat, #=BigFloat,=# Float16,
	Float32, Float64, Integer, Bool, Signed, BigInt, Int128, Int16, Int32, Int64, Int8,
	Unsigned, UInt128, UInt16, UInt32, UInt64, UInt8]
	@test all(R -> R<:T, Base.return_types(T))
	end

(There are two deprecated BigFloat methods which I haven't touched that would fail this test.)

However, while that holds for the individual constructor methods, for unknown argument types, usually too many methods match, so that e.g. Float64(x::Any) (26 possible methods) will still only be inferred as Any. So while I don't expect these changes to be wrong, I am not really sure how useful they are. Maybe @timholy has some intuition about it from all his work on invalidations where he looked at a lot code with incomplete type information?

[timholy]

I think these are worth having even if they don't solve the problem for cases of abstract inference. But yes, I think we'll still have to annotate the return type in a few cases.

Some day it might be helpful to have a tool that goes through the source code and suggests type-annotations to remove, because all inference problems have been fixed. It would be much nicer to write such a tool if we had #31162 and similar links so that we could transition seamlessly from the type-inferred code to the source code, but there are other ways of achieving similar ends.

[JeffBezanson]

I think T is not defined here?

[martinholters]

D'Oh. I certainly did not intend the test to be that broken. Fix incoming.