I called them $x,y$ and the ratio $z.$ There are two infinite families that make up all coprime integer solutions to
$$ x^2 + y^2 + yz - zx = 0 $$
The first family is
$$ x=u^2 + uv \, , \; \; y = -u^2 + uv \, , \; \; z = u^2 + v^2 $$
For this family $$ \frac{xy(x+y)}{x-y} $$
is always an integer. There is, in particular, no need to investigate common factors anywhere.
The second family is
$$ x=2u^2 + 3uv \, , \; \; y = u^2 + uv \, , \; \; z = 5u^2 +4uv v^2 $$
Let me look at that again, I don't appear to have written it out
For this family
$$ \frac{xy(x+y)}{x-y} = \frac{u^2 (u+v)(2u+3v)(3u+4v)}{u+2v} $$
and this may not be an integer. It ought to be possible, and quick, to check the individual conditions, (pretending the the denominator is prime) and working out the possible patterns for $u,v.$ Note that we can take coprime $u,v$ and just multiply both by $u+2v,$ since the new denominator is $(u+2v)^2$ but the numerator has been multiplied by $(u+2v)^5$
BACKGROUND:
Solving a homogeneous quadratic equation in three variables over the integers.
Help solving $ax^2+by^2+cz^2+dxy+exz+fzy=0$ where $(x_0,y_0,z_0)$ is a known integral solution
Describe the rational points on $3x^2 + y^2 = 4$
Question about non-degenerate polynomials, and a proof
Restrictions on the [hypothetical] solutions of a ternary quadratic form
Ternary Quadratic Forms MY PROOF!!! isotropic ternary!!!
$(a^2+b^2-ab)(a^2+b^2-2ab)$ is a perfect square eisenstein
Solutions to a quadratic diophantine equation $17x^2 + 29y^2 = 41z^2$
Identifying relation between numbers based on equation relating them