How do I know that one of the above statements is false without attempting, and failing, to prove it many, many different ways?
You may simply make a truth table for each of these statements, although that's not the most economic way, it's still one of the most straightforward techniques. Let's take for instance the first one, namely,
$$x \land ( y \to z ) \equiv ( x \land y ) \to ( x \land z ),$$
which we'll call $\small\tt P_1$. We have:
$$
\begin{array}{|c|c|c|c|c|c|c|c|}\hline
{}
\small \color{green}{x} &\small \color{green}{y} &\small \color{green}{z} &\small y\to z &\small x\land(y\to z) &\small x\land y &\small x\land z &\small (x\land y)\to(x\land z) &\small \tt P_\color{black}{1} \\ \hline
{}
\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} \\ \hline
\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} \\ \hline
\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} \\ \hline
\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} \\ \hline
\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} \\ \hline
\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} \\ \hline
\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} \\ \hline
\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{#C00}{0} &\small \color{royalblue}{1} &\small \color{#C00}{0} \\ \hline
\end{array}
$$
Hence $\small\tt P_1$ is certainly not a tautology. The use of a truth table is unnecessary if we observe from the beginning that we do not have an equivalence when $x$ is false. You can do the same for the remaining ones,
$$\begin{align}
x \lor ( y \to z ) &\equiv ( x \lor y ) \to ( x \lor z ),
\\
( y \lor z ) \to x &\equiv ( y \to x ) \lor ( z \to x ),
\end{align}$$
which we will respectively denote as $\small\tt P_2$ and $\small\tt P_3$. You'll see that $\small\tt P_2$ is the only tautology.
Now, to show that this is indeed the case $-$ without relying on a truth table $-$ we'll start with the RHS,
$$\begin{align}
(y\to x)\lor(z\to x) & \equiv (\lnot y\lor x)\lor(\lnot z\lor x) \\
& \equiv (\lnot y\lor\lnot z)\lor x \\
& \equiv \lnot(y\land z)\lor x \\
& \equiv (y\land z)\to x. \tag*{$\small\square$}
\end{align}
$$
A practical application of intuitionistic logic comes up naturally in the study of modern algebraic geometry.
To be brief, algebraic geometry is the study of the geometry of solution sets to systems of polynomial equations. At some point, it becomes useful to study the sheaf of regular functions on such an object. This means that we are interested in rational functions defined on "open" subsets; here the notion of "open" is given by the Zariski topology, where we just want to impose some conditions that certain other polynomials are nonzero. More generally, we can have functions which look like rational functions only locally.
However, it then turns out that if you generalize to sheaves of sets (as opposed to the sheaves of abelian groups, or sheaves of modules over a sheaf of rings, that are usually studied in algebraic geometry), then that category of sheaves of sets can naturally be given what is known as an "internal logic", which gives a "type theory" which acts very close to usual mathematics. The big difference is that in general, excluded middle and double negation elimination do not hold in this internal logic.
The usefulness then comes in being able to use the internal language of a topos to define interesting objects in the category of sheaves of sets. Also, if you have any statement which is formally provable in the internal logic, then it will always hold on any sheaves of sets. That can sometimes be useful in clarifying what otherwise could be a confusing proof involving passing to refinement covers multiple times.
So, even if you believe that the universe containing the objects of algebraic geometry satisfies the law of excluded middle, you still get interesting models of intuitionistic logic which do not satisfy excluded middle, and for which theorems of this version of intuitionistic logic can give useful information in the original universe.
(Modern algebraic geometry also studies so-called "Grothendieck sites" which generalize the example of Zariski topology: the etale sites, the crystalline site, etc. You can also define sheaves of sets on these sites, and they also have a straightforward generalization of the internal logic.)
Best Answer
Here are proofs in a fairly standard natural deduction system for intuitionistic logic.
For (3), assume $(p\lor q)\&(p\lor r)$. In particular, we have $p\lor q$, so we can consider separately two cases.
Case 1: $p$. Then the desired conclusion $p\lor(q\&r)$ follows immediately.
Case 2: $q$. In this case, we go back to the initial assumption to obtain $p\lor r$. So we can consider two subcases.
Subcase 2a: $p$. Again the desired conclusion follows immediately.
Subcase 2b: $r$. Combine this with $q$, which we have because we're in Case 2, to get $q\&r$ and thus the desired conclusion $p\lor(q\&r)$.
We got the desired conclusion in all cases and subcases, so the proof is complete.
For (4), assume $p\&(q\lor r)$. So we know $p$ and we know $q\lor r$. The latter lets us consider two cases separately.
Case 1: $q$. So we have $p\&q$, and therefore $(p\&q)\lor(p\&r)$, as required.
Case 2: $r$. So we have $p\&r$, and therefore $(p\&q)\lor(p\&r)$, as required.