Add "recipes" to look for certain text strings

[timmyte]

Add recipes to fetch certain strings easily.

Known strings:

• HURRAY

We might also have to use that for quantities, that are not yet in the JSON files.

Sort these recipes maybe in to groups of related targets.

[timmyte]

Keyword collection:
For TD-DFT/CIS convergence :
*** CONVERGENCE OF RESIDUAL NORM REACHED ***
For CCSD convergence:
--- The Coupled-Cluster iterations have converged ---
For CASSCF convergence:

                     ---- THE CAS-SCF ENERGY   HAS CONVERGED ----
                     ---- THE CAS-SCF GRADIENT HAS CONVERGED ----
                            --- FINALIZING ORBITALS ---
                    ---- DOING ONE FINAL ITERATION FOR PRINTING ----

I will put some examples here:
01 Failed Geometry Optimization
Input:

!noautostart
!wb97x-3c
!def2-tzvp
!opt
%pal
    nprocs 4
end

%geom
	MaxIter 2
end

*xyz 0 1
O    -3.56626    1.77639    0.0
H    -2.59626    1.77639    0.0
H    -3.88959    1.3604    -0.81444
*

Output Error:

    ----------------------------------------------------------------------------
                                  WARNING !!!
       The optimization did not converge but reached the maximum number of
       optimization cycles.
       Please check your results very carefully.
    ----------------------------------------------------------------------------

No other error or warning is printed. Missing HURRAY works as well.
(small side note: wb97x-3c should not be employed with a def2 basis since it comes with its own predefined basis set as each -3c method does!)

[timmyte]

02 Failed CC convergence
Input:

!noautostart
!ccsd(t)
%pal
    nprocs 4
end

%mdci
maxiter 2
end

*xyz 0 1
O    -3.56626    1.77639    0.0
H    -2.59626    1.77639    0.0
H    -3.88959    1.3604    -0.81444
*

Output Error:

               --- The Coupled-Cluster iterations have NOT converged ---

Otherwise no keywords/Errors are printed
For DLPNO-CCSD the keyword for not converging the CC equations is the same
The keyword for converged CC equations is:

               --- The Coupled-Cluster iterations have converged ---

[timmyte]

03 Failed CP-SCF for frequencies
Input:

!noautostart
!def2-tzvp
!tpss
!freq
%pal
    nprocs 4
end

%method
	Z_MaxIter 2
end

*xyz 0 1
O    -3.56626    1.77639    0.0
H    -2.59626    1.77639    0.0
H    -3.88959    1.3604    -0.81444
*

Output:

-------------------
SHARK CP-SCF DRIVER
-------------------

Dimension of the orbital basis          ...     43
Dimension of the CPSCF-problem          ...    190
Number of operators                     ...      1
Max. number of iterations               ...      2
Convergence Tolerance                   ... 1.0e-03
Number of perturbations                 ...      9
Perturbation type                       ... REAL

----------------------------
POPLE LINEAR EQUATION SOLVER
----------------------------

  ITERATION   0:   ||err||_max =   2.2861e-01 (   0.0 sec   0/  9 done)
  ITERATION   1:   ||err||_max =   1.5387e-02 (   0.0 sec   0/  9 done)

CP-SCF equations solved in   0.0 sec
Response densities calculated in   0.0 sec

No Error is thrown at all! The frequencies are just bad. The ORCA developers are working on this. Therefore the behavior will change and we just might just wait before adding a workaround (checking if 9/ 9 done etc). Just added this here not to forget it.

[timmyte]

04 Failed CIS Equations
Input:

!b3lyp
!def2-tzvp
!cpcm(hexane)

%tddft
    nroots 31
    maxiter 1
end

*xyz 0 1
O    0.0    0.0    0.0
H    0.0    -0.8    0.5
H    0.0    0.8    0.5
*

Output error:

[file orca_cis/cis_solve_cis.cpp, line 305]: CIS/TDA-DFT did not converge! Maybe try raising the number of iterations or use SOLVER= BHP22.


ORCA finished by error termination in CIS
Calling Command: orca_cis job.cisinp.tmp 
[file orca_tools/qcmsg.cpp, line 394]: 
  .... aborting the run

[timmyte]

@haneug
Add example notebook that shows how `recipes can be generated by the users with the available grepper