Fix source statistics

[dkachuma]

Fixes: #4010

This PR addresses issues regarding the collection and merge of source flux statistics.

Changes Made

• Time Advancement: Ensures that the time is properly advanced when a new collection is completed.
• Statistics Reset: Forces a reset of statistics prior to merging to prevent stale or cumulative data from corrupting the merge output.
• CSV File Handling: Updates the file-writing behaviour so the CSV file is appended to rather than overwritten each time, preserving historical collection data.
• Fix the element count to exclude ghost cells.

[MelReyCG]

just a few comments requests

[MelReyCG]

I was thinking, if a process is not involved in this computation, could it be excluded in some way? (if elementcount is 0 as an example)

[dkachuma]

@copilot Review and suggest code comments

[Copilot]

@copilot Review and suggest code comments

Added explanatory comments at both max() call sites (commit a7f0894):

• finalizePeriod() (line ~382): explains that MPI ranks without cells in the flux region can bypass gatherTimeStepStats() and lag behind, so MpiWrapper::max is used to ensure all ranks converge on the most advanced period start time.
• combine() (line ~404): explains that the same lag can occur across individual flux stats, so LvArray::math::max picks the most advanced time when merging two WrappedStats.

[dkachuma]

Added explanatory comments at both max() call sites (commit a7f0894):

@copilot This is OK but I expected to have to approve the commit.

[Copilot]

Added explanatory comments at both max() call sites (commit a7f0894):

@copilot This is OK but I expected to h...

As a coding agent, I commit and push changes directly when making edits — there's no approval step in between. If you'd prefer to review proposed changes before they're committed, you can describe what you'd like in a comment and I'll outline the changes first before applying them.