Fix bug in normalization of tally results with no_reduce

[paulromano]

Description

I tried using the little-known settings.no_reduce option recently (instructs each MPI rank to treat its own tally contributions as a single realization rather than reducing tallies across all MPI ranks) and found that it was broken due to a normalization error. This PR fixes the bug and adds a test so that it doesn't break in the future.

Checklist

• I have performed a self-review of my own code
• I have run clang-format (version 15) on any C++ source files (if applicable)
• I have followed the style guidelines for Python source files (if applicable)
• I have made corresponding changes to the documentation (if applicable)
• I have added tests that prove my fix is effective or that my feature works (if applicable)

[pshriwise]

This looks good to me, but I'll let it sit for a couple of days so others can have a look as well.