I see that on the page for "Bose-Einstein condensate", in the section "Derivation". I don't see it in the "Derivation" section on the "Bose-Einstein statistics" page, which is where I was hoping to see an explanation (of what it means that it showed up here).
But this page has some of the desired info: https://en.wikipedia.org/wiki/Polylogarithm#Integral_represe...
"For ''z'' = 1 the polylogarithm reduces to the Riemann zeta function:
\operatorname{Li}_s(1) = \zeta(s) \qquad (\operatorname{Re}(s)>1)"
And
"The polylogarithm can be expressed in term of the integral of the Bose-Einstein distribution:
\operatorname{Li}_{s}(z) = {1 \over \Gamma(s)} \int_0^\infty {t^{s-1} \over e^t/z-1} \,dt"
FWIW, I think the reason it does not show up on the "Bose-Einstein statistics" page: The zeta function is not so relevant to describing how the bosons populate states at particular energies, but zeta is relevant once you start summing/integrating over those populations (eg calculating expectation values.)
Unless I'm mistaken, it [e: meaning the page on condensates] does: Look at the "Derivation" section, and you'll see that g_3/2(f) is eventually manipulated into zeta.
[Or maybe I'm misunderstanding your comment.]