Fix new test case to work on (some?) big-endian architectures.

Use of pack("L") gets around the basic endian problem, but it doesn't
deal with the fact that the order of the bitfields within the struct
may differ.  This patch fixes it to work with gcc on NetBSD/macppc,
but I wonder whether that will be enough --- in principle, there
could be four different combinations of bitpatterns needed here.

Discussion: https://postgr.es/m/1650745.1679513221@sss.pgh.pa.us

Author: tglsfdc

src/bin/pg_amcheck/t/004_verify_heapam.pl

@@ -630,7 +630,7 @@ sub header
 		die "offnum $offnum should be a redirect" if defined $tup;
 		sysseek($file, 92, 0) or BAIL_OUT("sysseek failed: $!");
 		syswrite($file,
-				 pack("L", $ENDIANNESS eq 'little' ? 0x00010011 : 0x11000100))
+				 pack("L", $ENDIANNESS eq 'little' ? 0x00010011 : 0x00230000))
 			or BAIL_OUT("syswrite failed: $!");
 		push @expected,
 		  qr/${header}redirected line pointer points to another redirected line pointer at offset \d+/;
@@ -647,7 +647,7 @@ sub header
 		# rewrite line pointer with lp.off = 25, lp_flags = 2, lp_len = 0
 		sysseek($file, 108, 0) or BAIL_OUT("sysseek failed: $!");
 		syswrite($file,
-				 pack("L", $ENDIANNESS eq 'little' ? 0x00010019 : 0x19000100))
+				 pack("L", $ENDIANNESS eq 'little' ? 0x00010019 : 0x00330000))
 			or BAIL_OUT("syswrite failed: $!");
 		push @expected,
 		  qr/${header}redirect line pointer points to offset \d+, but offset \d+ also points there/;