Normally thermals used by glider pilots are driven by heating of the ground. 

> Thermals rise from the ground due to uneven heating from the sun, then
> stop rising once they start condensing into cloud.

That's not quite the right "spin" on things.  The top of the cumulus cloud does mark the point where the thermal has "run out of steam" and can go no higher.  But the fact that the thermal has reached the point where its water vapor has started condensing into cloud does not have any sort of damping influence on the thermal.  In fact, the condensation process actually *releases* heat into the air that helps the thermal rise higher than it otherwise would have.  This is due to the <a href="http://ww2010.atmos.uiuc.edu/(Gh)/wwhlpr/hurricane_latent_heat.rxml#:~:text=Latent%20heat%20is%20simply%20heat,around%20this%20condensation%20then%20warms.">"latent heat of condensation"</a>.  This plays an important role in the formation of thunderstorms and hurricanes.

> I often see multiple layers of clouds at the same time, are these
> higher layers formed by thermals or some other process?

Normally, by some other process.  These higher clouds do not normally indicate the presence of thermal updrafts that would be useable by glider pilots.  

However, one can sometimes notice that high-level stratus clouds can be heated by the sun enough to cause little cumulus-like "turrets" to form on their upper surfaces.  While not indicating lift that would be useable to glider pilots, this is still an interesting phenomenon.

<a href="https://en.wikipedia.org/wiki/Castellanus">This link on "Castellanus" clouds</a> may be of interest.

The sky is complicated!

PS I'm assuming that you are not envisioning that a glider pilot might actually be *towed up into a cloud* and then procede to fly on instruments to gain additional altitude inside the cloud after release from tow.  (As far as I know, this is never done, and would be illegal in some (<a href="https://aviation.stackexchange.com/q/84107/34686">but not all</a>) countries.)  If such a technique *were* envisioned, the scope of the question of which clouds indicate "useable" lift, would be considerably broadened!  See in particular the link on Castellanus clouds above-- it appears that these clouds contain rising air, but the rising air often does not extend to *below* the level of the cloud.