[PWGHF] Lc decay time: fix definition and provide additional info

[lubynets]

1. Fix definition of MC-gen decay time
In the current version of code the decay time of MC generated $\Lambda_{\mathrm{c}}^+$ particle is defined as $t^{\mathrm{gen}} = t / \gamma$, where $t$ is the lab-frame production time of one of the daughters, and $\gamma$ is the Lorentz factor of $\Lambda_{\mathrm{c}}^+$.
This PR proposes to define the decay time of MC generated $\Lambda_{\mathrm{c}}^+$ particle as $t^{\mathrm{kin}} = \frac{Lm}{pc}$, where $L$ and $p$ are MC-defined decay length and momentum of the mother particle.
For prompt $\Lambda_{\mathrm{c}}^+$ there should be no difference between these two definitions. The existing difference (left panels of the attached picture) are probably related to numeric artifacts and/or neglecting the curvature of $\Lambda_{\mathrm{c}}^+$ trajectory (?)
For nonprompt $\Lambda_{\mathrm{c}}^+$ the difference is more pronounced (right panels of the picture). The motivation to use the $t^{\mathrm{kin}}$ definition is that we are interested not in real proper decay time of nonprompt particle, but in its proxy, the quantity which we would obtain in data-driven way, if this particle was prompt (since for reconstructed decay candidates both in MC and data we use exactly this definition).

2. The THnSparse with reconstructed MC candidates is extended with yet another axis, containing decay time of matched generated particle (for obtaining the response matrix in order to enable unfolding / forward-folding).
3. The name lifetime in code variables is replaced with decay time, since the former is the property of particle species (PDG constant), while the latter is the property of particular particle instance.
4. Code cleaning
-- unneeded template typename replaced with explicit specification of parameter's type;
-- some variables are named less cryptic;
-- fixed clang-tidy errors.

[github-actions]

O2 linter results: ❌ 0 errors, ⚠️ 0 warnings, 🔕 0 disabled