The proof of Bool.noConfusion seems to be incorrect in Lean

[ColorlessBoy]

Prerequisites

Please put an X between the brackets as you perform the following steps:

• Check that your issue is not already filed:
  https://github.com/leanprover/lean4/issues
• Reduce the issue to a minimal, self-contained, reproducible test case.
  Avoid dependencies to Mathlib or Batteries.
• Test your test case against the latest nightly release, for example on
  https://live.lean-lang.org/#project=lean-nightly
  (You can also use the settings there to switch to “Lean nightly”)

Description

The proof of Bool.noConfusion seems to be incorrect in Lean.

Context

The bug was noticed by @ColorlessBoy on zulip.

Steps to Reproduce

I used the command #print Bool.noConfusion to inspect the proof of Bool.noConfusion:

@[reducible] protected def Bool.noConfusion.{u} : {P : Sort u} →
  {v1 v2 : Bool} → v1 = v2 → Bool.noConfusionType P v1 v2 :=
fun {P} {v1 v2} h12 =>
  Eq.ndrec (motive := fun a => v1 = a → Bool.noConfusionType P v1 a)
    (fun h11 => Bool.casesOn v1 (fun a => a) fun a => a) h12 h12

Upon checking, I found that the proof is incorrect because:
• There is a type mismatch in the first fun a => a, which is expected to have type Bool.noConfusionType P false v1 : Sort u;
• Similarly, there is a type mismatch in the second fun a => a, which is expected to have type Bool.noConfusionType P true v1 : Sort u.

Why does Lean 4 generate this incorrect proof for Bool.noConfusion?

Versions

Lean (version 4.15.0-rc1, arm64-apple-darwin23.6.0, commit ffac974dba79, Release)

Additional Information

Impact

Add 👍 to issues you consider important. If others are impacted by this issue, please ask them to add 👍 to it.