If $b\mid c$, prove that $\gcd(a,b)=\gcd(a+c,b)$. Hint let $d=\gcd(a,b)$ and $e=\gcd(a+c,b)$ and show that $d\mid e$ and $e\mid d$.
I already showed that $e\mid d$ but I can't figure out how to show $d\mid e$. I've tried substituting in for $d$ and $e$ but nothing seems to work. I think I'm not understanding a key fact.
You can't assume $x = nl$ and $y = mnl + ml$. Instead your argument should proceed as follows:
If $d = ax + by$ and $e = ax' + by'$ then this shows that $d \le e$ because the gcd of $a$ and $b$ is the smallest positive integer of the form $a \cdot \text{something} + b \cdot\text{something}$. With your method, you won't need this, but if you work a bit harder, you can show that $d \mid e$:
Likewise, you can write
$$d = ax + by = (a + c - c)x + by = (a + c)x + b(y - mx),$$
and since $e$ is the smallest postive integer of the form $(a + c) \cdot \text{something} + b \cdot\text{something}$, this shows that $e \le d$.
However, you don't have to use Bézout's theorem at all. The greatest common divisor $d$ of $a$ and $b$ is by definition:
A common divisor (i.e. $d \mid a$ and $d \mid b$)
The greatest divisor with this property (i.e. if $d' \mid a$ and $d' \mid b$ then $d' \le d$ [it is also true that $d' \mid d$])
Using the definition directly, you can show that if $e = \gcd(a + c, b)$ then $e \mid b$ (obvious) and $e \mid c$ (since $e \mid b \mid c$) therefore $e \mid a = (a + c) - c$ (since $e \mid a + c$ and $e \mid c$). Then, by (2.) it follows that $e \le d$.
Likewise, if you show that $d \mid b$ and $d \mid a + c$, (2.) says that $d \le e$.