Remember that a rational function is continuous across its domain. That is, it is continuous wherever it is defined. We can see that the function $f(x)$ is undefined whenever the denominator is $0$ as division by zero is undefined. You claim in your question that $f(x)$ is discontinuous at $x=-2$; I assure you that it is not. Because $f(-2)$ is defined, $x=-2$ is in the domain of this function and is therefore continuous there.
$$f(-2)=\frac{(-2)^2-3(-2)+2}{(-2)^2+(-2)-6}=\frac{4+6+2}{4-2-6}=\frac{12}{-4}=-3$$
We can simplify $f(x)$ only if $x\ne2$. This is because the denominator will become $0$ if $x=2$.
$$f(x)=\frac{x^2-3x+2}{x^2+x-6}=\frac{(x-2)(x-1)}{(x-2)(x+3)}$$
Notice what happens if we let $x=2$.
$$f(2)=\frac{((2)-2)((2)-1)}{((2)-2)((2)+3)}=\frac{0\cdot1}{0\cdot5}=\frac{0}{0}$$
This is undefined (it is also an indeterminate form). Notice how you obtained $1/5$ here. By simply removing $(x-2)$ from both the numerator and denominator, you essentially removed $0/0$ when $x=2$. You cannot do this. You will obtain confusing and erroneous results when you mistakenly divide by zero. This is the main point of your confusion here. Now, let us determine what type of discontinuity exists at $x=2$ as the problem asks. We know that $f(2)$ is undefined, but we can evaluate the limit as $x\to2$ to see what is happening there. We can remove $(x-2)$ when evaluating the limit because $x$ is only approaching $2$; it will not equal it, so we won't have a problem. That is what a limit implies. We can use direct substitution.
$$\lim\limits_{x\to2}(f(x))=\frac{(2)-1}{(2)+3}=\frac{1}{5}$$
Because $f(2)$ is undefined and the limit as $x\to2$ exists, we can then say that there exists a removable discontinuity at $x=2$. This is sometimes called a "hole" because of a missing point in the graph of the function where $(2,f(2))$ should be. If $f(x)$ were to be continuous at $x=2$, it would need to be defined there and equal to this limit. Thus, $f(2)$ would need to be equal to $1/5$. In all, the function $f(x)$ is continuous on $(-\infty,-3)\cup(-3,2)\cup(2,\infty)$ as that is the domain of the function. Alternatively, you could say there is a removable discontinuity at $x=2$ because we could "remove" the expression $(x-2)$ from the function as it would have lead to a discontinuity at a certain value (if $x=2$). This intuition, of course, is informal and simply pedagogical. The same cannot be said for $x=-3$, however. The function is also discontinuous there, but it is a different type of discontinuity—an infinite discontinuity.