I am trying to answer this question from Schaum's Calculus. Can someone please let me know if my answer looks kosher?
If we substitute $y=mx\pm \sqrt{a^2m^2+b^2}$ into the equation $b^2x^2+a^2y^2=a^2b^2,$ we obtain
$$b^2x^2+a^2(mx\pm \sqrt{a^2m^2+b^2})^2=a^2b^2,$$
which can be algebraically manipulated to arrive at
$$mx\pm \sqrt{a^2m^2+b^2}=\pm \sqrt{\frac{a^2b^2-b^2x^2}{a^2}}.$$
Solving the ellipse equation for $y$ yields
$$y=\pm \sqrt{\frac{a^2b^2-b^2x^2}{a^2}}.$$
Thus, I get that
$$y=mx\pm \sqrt{a^2m^2+b^2}.$$
Have I used any tautologies here, or does it look good? Thanks for your help!
Here is a way:
Let the tangent be $y=mx+c$. Put this in the ellipse's equation and then set the discriminant of a quadratic in $x$ equal to zero (Since the tangent and ellipse intersect at only one point).
Another way to go forward would be using the derivative. I think you did a circular reasoning to prove the statement...Nowhere you used the formula of the tangent ...You might as well use $y=\textit{Anything}$ and still arrive at the same result...
Edit:
Putting $y=mx+c$ into the ellipse's equation, we get $$b^2x^2+a^2(m^2x^2+2mxc+c^2)-a^2b^2=0$$ Rearranging for the quadratic in $x$
$$x^2(a^2m^2+b^2)+2mca^2x+c^2a^2-a^2b^2=0$$ Setting the discriminant zero and solving for $c$ we get
$c^2=a^2m^2+b^2$ And we are done.
Note that we are getting two values of $c$ which are both valid as there could be $2$ tangents with a given slope $m$ to a standard ellipse which can be seen from the graph.