DURACLOUD-1310: use number in chunkId when index number is possibly missing in a chunk manifest

[nwoodward]

This PR changes the file stitching process to handle cases where the index number is missing from entries in a chunk manifest.

---

JIRA Ticket: https://duracloud.atlassian.net/browse/DURACLOUD-1310

What does this Pull Request do?

Adds a comparison between the index number and the number at the end of each chunkId.

How should this be tested?

1. Use the SyncTool to upload a large file that gets chunked to a DuraCloud space.
2. Modify only the dura-manifest file in the space by removing the index=X tag from each tag.
3. Use the production version of the RetrievalTool to retrieve the file. It will fail with a checksum mismatch error.
4. Use this snapshot version of the RetrievalTool to retrieve the file.

Interested parties

Tag @duracloud/committers

[bbranan]

I've tested this and it works as expected, solving the problem with missing index entries in the manifest xml. The code looks good as well except that there is currently more logging than is really needed long term.

In the case where the manifest index and parsed index match, I see no need to log as everything is working as expected. I can see including an info-level log when they don't match, but perhaps updating the text to be more understandable by the end-user that is likely to be seeing the message. Maybe something like:
log.info("The entry in the chunk manifest for chunk {} is missing an index field; using the index from the filename instead.", entry.getChunkId());