empty band char in extrema output

I was running phase diagram batch across zero-gap and some good portion of the calculations end up with incorrect band alignment.

<cmdargs n_args="77">/home/elvira/kdotpy_venv/bin/kdotpy 2d norb 8 cpus 11 gpus 11 bia noax broadening 2 gauss sqrt erange -100 50 split 0.01 msubst CdTe mlayer HgCdTe 70% HgCdTe 0.048 HgCdTe 70% llayer 11 10.4 13 zres 0.2 matparam 'HgTe:P=sqrt(21000*hbarm0);CdTe:P=sqrt(21000*hbarm0);HgCdTe:P=sqrt(21000*hbarm0)' matparam 'HgCdTe:Eg=leq(x-0.1)*(-329+1935*x)+gtr(x-0.1)*(-303*(1-x)+1606*x-132.*x*(1-x)+(0.495*(1-x)-0.325*x-0.393*x*(1-x))*T**2/(11.0*(1-x)+78.7*x+T))' matparam 'CdTe:Evoff=-620*(Eg-Eg0)/(1606+303);HgCdTe:Evoff=-620*(Eg-Eg0)/(1606+303)' nll 10 llfull k 0 0.6 / 0.01 kphi 45 extrema neig 50 targetenergy -10 carrdens -0.0012 0.0012 / 0.0002 transitions llobs dos legend char obs llavg plotfz verbose config 'band_align_exp=2;lattice_regularization=false;batch_stderr_extension=kperr;batch_stdout_extension=kpout;dos_energy_points=1000;diag_solver=cupy_eigsh;diag_solver_cupy_dtype=double;transitions_plot=false;transitions_spectra=false;transitions_min_amplitude=1e9' temp 4.2 outdir /home/elvira/data//kdisp_EgWx_bianoax/d_10.4_Cd_0.048/EP_21000_VBO_-620 out _d_10.4_Cd_0.048_EP_21000_VBO_-620</cmdargs>

Once you get faulty alignment extrema determination is also confused, leaving band character column empty, which can further interfere with processing scripts.

I suggest we write some string there like "undefined" instead of leaving the field empty