Why is the worst case very useful for algorithm efficiency analysis? $(30$ points$)$

When all elements in the input array A are the same, is it the best case or worst case for the INSERTION$-$SORT function below? Why? $(35$ points$)$

INSERTION$-$SORT $(A)$

$1$ for $j=2toA.$length

$2$ key $=A[j]$

$3\frac{\frac{?}{?}}{\text{I}}$nsert $A[j]$ into the sorted

sequence $A[1$dotsj$-1].$

$4,i=j-1$

$5$ while $i>0$ and $A[i]>$ key

$6,A[i+1]=A[i]$

$7,i=i-1$

$8,A[i+1]=$ key

Between the following two input arrays, $A=[1,2,3,4,5,6,7,8]$ and $,6,5,4,3,2,1,$ which requires more steps to run the MERGE$-$SORT function below? Why? $(35$ points$)$

Merge$-$Sort$(A,p,r)$

$ifp$