Seperate data uploading from job summission for DS2 cloud training and add support for multiple shards uploading.

Summary:
1. Add missing is_local argument (when set False, use pserver).
2. Add exception thrown if cp failed.
3. Add cloud mkdir if not cloud path for uploading does not exist.
4. Fix a bug using common path ./local_manifest for all nodes. (convert to /local_manifest)
5. Refine coding style.