Equality relation for TruthDicts

The following code is wrong.

```julia
my_model = randmodel(42, 5, 10, [Atom("p"), Atom("q")], BooleanAlgebra())
my_model.assignment == deepcopy(my_model.assignment)
```

The assignment within `my_model::KripkeModel` is a `TruthDict`, but it happens that a truth dict is never equal to itself.

Look at the TruthDict type definition and its equality method (likely == or isequal). The issue shows a failing equality test between a TruthDict and its deep copy. Start by finding where TruthDict is defined (search in src/), examine the current equality implementation, and write a test to reproduce the problem. Check if the issue is in how the dict's internal structure or keys are compared.

索引モデルが issue の本文から書いたものです。