Fix intermittency in the sys.fs.fusefs.mknod.main test

In the Mknod.parent_inode test case, the kernel sends an extra
FUSE_FORGET message.  But because it gets sent asynchronously with the
failing syscall, it doesn't always get received before the test ends.
So we never setup an expectation for it.  And 90+% of the time the test
would exit successfully.

Fix the intermittency by always waiting to receive the FUSE_FORGET
message.

MFC after:	2 weeks
Sponsored by:	Axcient

tests/sys/fs/fusefs/mknod.cc

@@ -32,6 +32,7 @@ extern "C" {
 #include <fcntl.h>
 #include <sys/socket.h>
 #include <sys/un.h>
+#include <semaphore.h>
 }
 
 #include "mockfs.hh"
@@ -255,14 +256,18 @@ TEST_F(Mknod, parent_inode)
 	const char RELPATH[] = "some_node";
 	mode_t mode = S_IFSOCK | 0755;
 	struct sockaddr_un sa;
+	sem_t sem;
 	int fd;
 	dev_t rdev = -1;	/* Really it's a don't care */
 	uint64_t ino = 42;
 
+	ASSERT_EQ(0, sem_init(&sem, 0, 0)) << strerror(errno);
+
 	expect_lookup(PPATH, ino, S_IFDIR | 0755, 0, 1);
 	EXPECT_LOOKUP(ino, RELPATH)
 	.WillOnce(Invoke(ReturnErrno(ENOENT)));
 	expect_mknod(ino, RELPATH, ino, mode, rdev);
+	expect_forget(ino, 1, &sem);
 
 	fd = socket(AF_UNIX, SOCK_STREAM, 0);
 	ASSERT_LE(0, fd) << strerror(errno);
@@ -273,6 +278,8 @@ TEST_F(Mknod, parent_inode)
 	ASSERT_EQ(EIO, errno);
 
 	leak(fd);
+	sem_wait(&sem);
+	sem_destroy(&sem);
 }
 
 /*