G4ParticleMessenger: fix incorrect GetCurrentValue implementation for G4ParticleMessenger

[veprbl]

This fixes a following bug:

Idle> /particle/select lambda
Idle> /particle/property/dump

--- G4ParticleDefinition ---
 Particle Name : lambda
 PDG particle code : 3122 [PDG anti-particle code: -3122]
 Mass [GeV/c2] : 1.11568     Width : 2.501e-15
 Lifetime [nsec] : 0.2631
 Charge [e]: 0
 Spin : 1/2
 Parity : 1
 Charge conjugation : 0
 Isospin : (I,Iz): (0/2 , 0/2 ) 
 GParity : 0
 MagneticMoment [MeV/T] : -1.93245e-14
 Quark contents     (d,u,s,c,b,t) : 1, 1, 1, 0, 0, 0
 AntiQuark contents               : 0, 0, 0, 0, 0, 0
 Lepton number : 0 Baryon number : 1
 Particle type : baryon [lambda]
G4DecayTable:  lambda
0:  BR:  0.639  [Phase Space]   :   proton pi-
1:  BR:  0.358  [Phase Space]   :   neutron pi0

Idle> /particle/property/decay/dump
Particle is not selected yet !! Command ignored.

The issue is that currentParticle is always nullptr, so when G4DecayTableMessenger tries to use GetCurrentStringValue with reGet=true default, it can never get a non-null particle pointer.

[drbenmorgan]

Many thanks for the patch, this is now integrated in the upstream development, and will be in the upcoming 11.3 release and patches to earlier versions!