rebase -r: do not write MERGE_HEAD unless needed

When we detect that a `merge` can be skipped because the merged commit
is already an ancestor of HEAD, we do not need to commit, therefore
writing the MERGE_HEAD file is useless.

It is actually worse than useless: a subsequent `git commit` will pick
it up and think that we want to merge that commit, still.

To avoid that, move the code that writes the MERGE_HEAD file to a
location where we already know that the `merge` cannot be skipped.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

sequencer.c

@@ -3191,10 +3191,6 @@ static int do_merge(struct commit *commit, const char *arg, int arg_len,
 	}
 
 	merge_commit = to_merge->item;
-	write_message(oid_to_hex(&merge_commit->object.oid), GIT_SHA1_HEXSZ,
-		      git_path_merge_head(the_repository), 0);
-	write_message("no-ff", 5, git_path_merge_mode(the_repository), 0);
-
 	bases = get_merge_bases(head_commit, merge_commit);
 	if (bases && oideq(&merge_commit->object.oid,
 			   &bases->item->object.oid)) {
@@ -3203,6 +3199,10 @@ static int do_merge(struct commit *commit, const char *arg, int arg_len,
 		goto leave_merge;
 	}
 
+	write_message(oid_to_hex(&merge_commit->object.oid), GIT_SHA1_HEXSZ,
+		      git_path_merge_head(the_repository), 0);
+	write_message("no-ff", 5, git_path_merge_mode(the_repository), 0);
+
 	for (j = bases; j; j = j->next)
 		commit_list_insert(j->item, &reversed);
 	free_commit_list(bases);

t/t3430-rebase-merges.sh

@@ -396,7 +396,7 @@ test_expect_success 'with --autosquash and --exec' '
 	grep "G: +G" actual
 '
 
-test_expect_failure '--continue after resolving conflicts after a merge' '
+test_expect_success '--continue after resolving conflicts after a merge' '
 	git checkout -b already-has-g E &&
 	git cherry-pick E..G &&
 	test_commit H2 &&