No, the experiment is not setup with the cat as the observer. That would be a different experiment and the cat would be measuring you and everything else including the inside of the box. Read Wigner's friend on this.
QM works like this. An experiment is done and the results are in. This is worked out classically then you calculate what the quantum wavefunction was. So there are two parts. The experiment is done, a measurement happens. They you figure out the wavefunction math structure. So you can't ask the question about the cat in the box before you do the measurement. It is not allowed to consider the cat before the final measurement is made i.e. opening the box and looking inside. So this time period when you can't ask any questions is called the superposition. So all these conversations about the state of the cat before the box is opened are not allowed. That is how all these people get the concept wrong. Whether or not the cat is alive or dead before the box is opened is not a valid question. Just like the electron and double slit. You can't ask the question about which slit the electron went through. If you do ask the question about which slit then you have just changed the whole experiment to a measurement that happened at the slit. So the original measurement at the screen wasn't done and the whole experiment has changed.

There is more to it than perturbing the particle because entangled particles show correlations even when light years apart. There spins are correlated more than possible with any hidden variable theory. You are speaking in terms of hidden variables that get changed when you measure something but QM is deeper than that. Check Bells inequality.

But QM comes in two steps, the wavefunction and squaring the wavefunction at the time of measurement. So the measurement happens. It is part of QM for the measurement to happen. Then you calculate what a wavefunction would have been. There in nothing that reaches a threshold and triggers a "collapse." The measurement is classical and the wavefunction is quantum. So think of the experiment being set up and carried out and you get results. It has to happen this way. If there is no measurement then you don't calculate a wavefunction. We know the measurements happen all the time but we never see a wavefunction and by definition we can't see it until the point of measurement and the measurement always happens or you don't have an experiment.

No. There is no collapse of wavefunctions. An experiment is set up and a result is achieved. Then you figure out the math of the wavefunction. So there is always a measurement. It has nothing to do with a conscious observer except you, the last one in the line. Read Wigner's friend and when you observe other people they are in superpositions. So other conscious people are in superpositions. The only that isn't is you.

The measurement is done first, that is the basis of QM and there are two parts. The squaring of the wavefunction is done when you measure and you are measuring everything around you all day long. Then you can figure out the mathematical wavefunctions that preceded the measurement. But you don't just get a measurement happening willy nilly any time. It is a classical procedure, making a measurement. That is how it works. There is no threshold number of particles you need or gravity waves triggering things. So start with the measurement, and since we know there was and still are on going measurements, all we know is that we can calculate a wavefunction after the fact. The wavefunctions don't get triggered to collapse , they are a mathematical procedure to back calculate.