Is it correct in saying that a particles size is it's rest energy,
No
and that particles don't actually have size (in the way you get different size objects)?
Not really.
Why do particles have discrete sizes, and there's not a continuous spectrum of particles varying in size?
They don't have discrete sizes.
OK, let me explain:
Fundamental particles are point particles by the Standard Model (String theory gives them a size, which seems to be somewhat correlated with mass, but I'm keeping String theory aside here).
Of course, we have fundamental particles of varying rest energies, so size is not related to mass.
Now, let's look at composite particles. Say, a proton. Protons are made of three quarks, along with some virtual gluons that are being exchanged. The size of the proton is thus the volume in which all its constituent particles are confined.
But, the constituent particles are not confined--there is a very slim chance of finding one of the quarks a few meters away{*}, due to the wave nature of the particles. So, we define size as the volume where you will find all the particles in <some arbitrary percentage>
of the time (where that arbit percentage will be somewhat like 99.999%)
The definition need not be "all the particles will be found here x% of the time"; we can have fundamentally similar, but different definitions. For example, as @dmckee noted, the charge radius of a nucleus is where the charge density drops to $e^{-1}$ of the central density.
Similarly, for an atom, you could define size via the boundary of the electron cloud. But, the electron cloud stretches to infinity, so you have to arbitrarily define size via a percentage(or something similar).
In fact, for an atom, going across a period, size decreases as mass increases (but not when going down a group). So, for composite particles, size and mass have no fixed relation.
*I'm not too sure of this, color confinement may prevent the quark from going so far without pair production.