I know that in s/// or m/// the /x will ignore white space ...

The /x modifier will allow you to add white space in your regular expression to make it more readable - it doesn't ignore white space in the text that you are trying to match. Example:

my $x = "foobar";
print "matches 1\n" if ($x =~ m/foo bar/x); # succeeds 
print "matches 2\n" if ($x =~ m/foo bar/); # fails
[download]

humm just had a thought ... how about running through the array and just removing all white space and dump into @clean, then do the same to $line just prior to the grep

This is a good idea. You are canonicalizing the input, mapping each line to a single representative so that you can just use string equality to test if the lines are "the same."

Another common perl-ish approach is to use a hash. This will avoid iterating through the the @clean array for each line in the second file:

my %hash;
while (<file1>) {
  my $clean = ... convert $_ into its canonical form ...
  $hash{$clean} = 1;
}
while (<file2>) {
  my $clean = ... convert $_ into its canonical form ...
  if ($hash{$clean}) {
    # line from file2 exists in file1
  } else {
    # line from file2 is not in file1
  }
}
[download]

This run time is essentially linear in the number of lines of each file.