subprocess: Fix memory leaks

I encountered this issue while running my code with Valgrind today.
Below is part of the Valgrind error message:

```
==1578139== 472 bytes in 1 blocks are still reachable in loss record 1 of 1
==1578139==    at 0x4848899: malloc (...)
==1578139==    by 0x4B3AF62: fdopen@@GLIBC_2.2.5 (...)
==1578139==    by 0x118B09: subprocess::Popen::execute_process() (...)
```

I noticed that a similar fix had been proposed by another contributor
previously. I did not mean to scoop their work, but merely hoping to fix
it sooner so other people don't get confused by it just as I did today.

Github-Pull: arun11299/cpp-subprocess#106
Rebased-From: 3afe581c1f22f106d59cf54b9b65251e6c554671

Author: HurryPeng

src/util/subprocess.h

@@ -1181,11 +1181,13 @@ inline void Popen::execute_process() noexcept(false)
     try {
       char err_buf[SP_MAX_ERR_BUF_SIZ] = {0,};
 
-      int read_bytes = util::read_atmost_n(
-                                  fdopen(err_rd_pipe, "r"),
-                                  err_buf,
-                                  SP_MAX_ERR_BUF_SIZ);
-      close(err_rd_pipe);
+      FILE* err_fp = fdopen(err_rd_pipe, "r");
+      if (!err_fp) {
+          close(err_rd_pipe);
+          throw OSError("fdopen failed", errno);
+      }
+      int read_bytes = util::read_atmost_n(err_fp, err_buf, SP_MAX_ERR_BUF_SIZ);
+      fclose(err_fp);
 
       if (read_bytes || strlen(err_buf)) {
         // Call waitpid to reap the child process