Incorrect output for `/test/basic/tree/3.wl` in model `sopl`

[julianayang777]

When running the cram tests for the model sopl, the expected output is wrong. But the when running with the binary wl the output is right:

=====================
	Ænima
=====================

Input file: test/basic/tree/3.wl
Execution mode: sopl

Outcome: Returned (Val Loc 0)
Outcome: Assertion violated, counter example:
				$_i : Int 5
				$_j : Int 0
				$_v : Int 4
				$_k : Int 5
Outcome: Assumption evaluated to false
Outcome: Assumption evaluated to false
Outcome: Assumption evaluated to false
Outcome: Assumption evaluated to false
Outcome: Assumption evaluated to false
Outcome: Assumption evaluated to false

Having the tests test/basic/tree/3.wl and test/basic/tree/4.wl and running the script test, is enough to trigger the bug. It might be something related with solver interactions.