add missing atom safeguards to FINDPRO and FINDNUC routines

Author: jayponder

README.md

@@ -1,12 +1,12 @@
 # Tinker: Software Tools for Molecular Design
 
-<I>Please note that the release of Tinker 25 in late February 2025 the code contains a number of new features, and is not fully backward compatible with prior versions of Tinker. In particular, do not use older versions of parameter files with current Tinker executables. Similarly, parameter files from the current Tinker may give errors or incorrect results when used with executables from earlier "Tinker 8" versions.</I>
+<I>Please note with the release of Tinker 25 in late February 2025 the code contains a number of new features, and is not fully backward compatible with prior versions of Tinker. In particular, do not use older versions of parameter files with current Tinker executables. Similarly, parameter files from the current Tinker may give errors or incorrect results when used with executables from earlier "Tinker 8" versions.</I>
 
 <H2><B>Introduction</B></H2>
 
 Tinker is a complete and general package for molecular mechanics and dynamics, with some special features for biopolymers. Tinker has the ability to use any of several common parameter sets, such as Amber (ff94, ff96, ff98, ff99, ff99SB, ff14SB, ff19SB), CHARMM (19, 22, 27, 36), Allinger MM (MM2-1991 and MM3-2000), OPLS (OPLS-UA, OPLS-AA, OPLS-AA/L), Merck Molecular Force Field (MMFF94, MMFF94s), Liam Dang's polarizable model, and the AMOEBA, AMOEBA+ and HIPPO polarizable atomic multipole force fields. Parameter sets for other widely-used force fields are under consideration for future releases.
 
-The Tinker software contains a variety of interesting algorithms such as: flexible implementation of atomic multipole-based electrostatics with explicit dipole polarizability, various continuum solvation treatments including several generalized Born (GB/SA) models, generalized Kirkwood implicit solvation for AMOEBA, an interface to APBS for Poisson-Boltzmann calculations, efficient truncated Newton (TNCG) local optimization, surface areas and volumes with derivatives, free energy calculations via the Bennett Acceptance Ratio (BAR) method, normal mode vibrational analysis, minimization in Cartesian, torsional or rigid body space, symplectic RESPA multiple time step integration for molecular dynamics, velocity Verlet stochastic dynamics, pairwise neighbor lists and splined spherical energy cutoff methods, particle mesh Ewald (PME) summation for partial charges and polarizable multipoles, a novel reaction field treatment of long range electrostatics, fast distance geometry metrization with better sampling than standard methods, Elber's reaction path algorithm, potential smoothing and search (PSS) methods for global optimization, Monte Carlo Minimization (MCM) for efficient potential surface scanning, tools for fitting charge, multipole and polarization models to QM-based electrostatic potentials and more....
+The Tinker software contains a variety of interesting algorithms such as: flexible implementation of atomic multipole-based electrostatics with explicit dipole polarizability, various continuum solvation treatments including several generalized Born (GB/SA) models, generalized Kirkwood implicit solvation for AMOEBA, an interface to APBS for Poisson-Boltzmann calculations, efficient truncated Newton (TNCG) local optimization, surface areas and volumes with derivatives, free energy calculations via the Bennett Acceptance Ratio (BAR) method, normal mode vibrational analysis, minimization in Cartesian, torsional or rigid body space, symplectic RESPA multiple time step integration for molecular dynamics, velocity Verlet stochastic dynamics, pairwise neighbor lists and splined spherical energy cutoff methods, particle mesh Ewald (PME) summation for partial charges and polarizable multipoles, a novel reaction field treatment of long range electrostatics, fast distance geometry metrization with better sampling than standard methods, Elber's reaction path algorithm, potential smoothing and search (PSS) methods for global optimization, Monte Carlo Minimization (MCM) for efficient potential surface scanning, tools for fitting charge, multipole and polarization models to QM-based electrostatic potentials and more.
 
 <H2><B>Current Release</B></H2>