Update handling of duplicate batches to avoid editing existing messages

[peterbroadhurst]

Proposed changes

Depends on #1649 going in first

If a FireFly node receives a batch a 2nd time, you can end up with messages that are:

• Marked pending (without a confirmed time)
• Have a message_confirmed event associated

Logically this should be impossible, but it happens because we overwrite the message state in the upsert logic, after the insert fails.

Easy to recreate this, by just using /operation/{opId}/retry to re-send a batch containing confirmed messages.

This PR changes the fallback logic, after the insert fails, to only insert the missing messages.

---

Types of changes

• Bug fix
• New feature added
• Documentation Update

---

• I have read the contributing guidelines.
• I have performed a self-review of my own code or work.
• I have commented my code, particularly in hard-to-understand areas.
• My changes generates no new warnings.
• My Pull Request title is in format < issue name > eg Added links in the documentation.
• I have added tests that prove my fix is effective or that my feature works.
• My changes have sufficient code coverage (unit, integration, e2e tests).

[awrichar]

Good find; thanks for fixing this.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.95%. Comparing base (490539e) to head (d934bcc).
Report is 6 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1647   +/-   ##
=======================================
  Coverage   99.95%   99.95%           
=======================================
  Files         337      337           
  Lines       29607    29632   +25     
=======================================
+ Hits        29595    29620   +25     
  Misses          8        8           
  Partials        4        4           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.