Improve CSTR model

[ZedongPeng]

Changes

• Remove the redundant boolean variables in the CSTR model.

[bernalde]

I would like you to review if this model can be solved by LOA as it looks similar to the changes that I made earlier with @AlbertLee125 that would fail in that.
More importantly, there are issues that are being resolved with the model right now so I would hold from merging these until we are clear in that front

[bernalde]

@AlbertLee125 any news?

[AlbertLee125]

if __name__ == "__main__":
    m = build_model(NT=5)
    subsolver = 'scip'
    pyo.SolverFactory("gdpopt.loa").solve(
        m,
        tee=True,
        nlp_solver=subsolver,
        mip_solver=subsolver,
        minlp_solver=subsolver,
        local_minlp_solver=subsolver,
    )

Running the reformed code with the given script. The model can be solved by LOA and the objective found by LOA is similar to the objective solved by GDP reformulation.