return type inference regression for accessing the parameters field of nonconrete subtypes of DataType

[nsajko]

julia> Base.infer_return_type((x -> x.parameters), Tuple{Type{Tuple{Int}}})
Core.SimpleVector

julia> Base.infer_return_type((x -> x.parameters), Tuple{Type{Tuple{Any}}})
Core.SimpleVector

julia> versioninfo()
Julia Version 1.11.3
Commit d63adeda50d (2025-01-21 19:42 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 8 × AMD Ryzen 3 5300U with Radeon Graphics
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver2)
Threads: 1 default, 0 interactive, 1 GC (on 8 virtual cores)

julia> Base.infer_return_type((x -> x.parameters), Tuple{Type{Tuple{Int}}})
Core.SimpleVector

julia> Base.infer_return_type((x -> x.parameters), Tuple{Type{Tuple{Any}}})
Any

julia> versioninfo()
Julia Version 1.13.0-DEV.173
Commit beb928bfec5 (2025-03-06 13:25 UTC)
Build Info:
  Official https://julialang.org release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 8 × AMD Ryzen 3 5300U with Radeon Graphics
  WORD_SIZE: 64
  LLVM: libLLVM-19.1.7 (ORCJIT, znver2)
  GC: Built with stock GC
Threads: 1 default, 1 interactive, 1 GC (on 8 virtual cores)

I'm guessing there's a isconcretetype(type) check somewhere that just needs a type isa DataType check?

[nsajko]

Bisected to #57275. EDIT: that PR is correct and a bugfix. However, it still seems like it may be possible to fix this inference regression by improving the "precision of the compiler model", accounting for the fact that DataType is the only one with a field named parameters, presumably in Compiler/src/tfuncs.jl.