fix(fossology): reset IN_WORK state on upload exception to prevent stuck clearingState

[Aman-Cool]

When sending a release to FOSSology, the handler writes IN_WORK to CouchDB before the upload attempt. If anything throws during the attachment fetch or network call, that state never gets cleaned up; leaving the release permanently stuck at SENT_TO_CLEARING_TOOL with no way to retry without admin intervention.

• Wrapped the step-handler dispatch in a try/catch/finally so updateFossologyProcessInRelease always runs, and any upload step interrupted mid-flight gets reset back to NEW.
• No new dependencies.

Issue: NA

Suggest Reviewer

@GMishx

How To Test?

1. Attach a source file to a release and configure FOSSology.
2. While clicking "Send to FOSSology", briefly make the CouchDB attachments DB unreachable (or revoke read access).
3. Before fix: release stays at SENT_TO_CLEARING_TOOL, every retry silently does nothing.
4. After fix: release resets to NEW_CLEARING, retry works normally.

No new automated tests added, the fix is in exception-path control flow.

Checklist

Must:

• All related issues are referenced in commit messages and in PR

[Aman-Cool]

Hey @GMishx, Found this while digging into the FOSSology flow; if the attachment fetch or any I/O throws between the pre-commit IN_WORK write and the actual upload, there's no cleanup path, so the release just silently hangs there forever. Pretty easy to hit during a CouchDB hiccup or rolling restart. The fix is just a try/catch/finally to make sure the state always gets corrected before we rethrow. Happy to add a test or tweak anything if needed!