t3903: reduce dependencies on previous tests

Skipping previous tests to work through only failing tests with
arguments like --run=4,122- causes some tests to fail because subdir
doesn't exist yet (it is created by a previous test; typically
"unstashing in a subdirectory"). Create it on demand for tests that need
it, but don't fail (-p) if the directory already exists.

Signed-off-by: D. Ben Knoble <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

Author: benknoble

t/t3903-stash.sh

@@ -895,6 +895,7 @@ test_expect_success 'branch: should not drop the stash if the apply fails' '
 
 test_expect_success 'apply: show same status as git status (relative to ./)' '
 	git stash clear &&
+	mkdir -p subdir &&
 	echo 1 >subdir/subfile1 &&
 	echo 2 >subdir/subfile2 &&
 	git add subdir/subfile1 &&
@@ -1327,6 +1328,7 @@ test_expect_success 'stash -k -- <pathspec> leaves unstaged files intact' '
 
 test_expect_success 'stash -- <subdir> leaves untracked files in subdir intact' '
 	git reset &&
+	mkdir -p subdir &&
 	>subdir/untracked &&
 	>subdir/tracked1 &&
 	>subdir/tracked2 &&
@@ -1343,6 +1345,7 @@ test_expect_success 'stash -- <subdir> leaves untracked files in subdir intact'
 
 test_expect_success 'stash -- <subdir> works with binary files' '
 	git reset &&
+	mkdir -p subdir &&
 	>subdir/untracked &&
 	>subdir/tracked &&
 	cp "$TEST_DIRECTORY"/test-binary-1.png subdir/tracked-binary &&