r.statistics: Fix uninitialized variable warning in o_var

[ShubhamDesai]

This pull request addresses the following warning identified by cppcheck.

Issue:
o_var.c:70:16: error: Uninitialized variable: count [uninitvar]
m_var(tab, count, &vari);

Changes made:
The issue was fixed by initializing the variable count to 0 at the beginning of the o_var function. This ensures that the variable has a defined value before it is used in the calculation.

[echoix]

Also, just a question like this, would it be possible to have multiple, or all, issues that have the exact same reasoning be fixed at once in one PR? (You can limit to a folder (like raster, vector, temporal, or group all of the modules together) if the amount of changes is too big). You can still leave out more complex fixes in a separate PR if you want. I think it would make better use of reviewer's time (as you will still need to wait about the same time to get a first answer, and we still need to read the context each time), but also CI time (2 x 1h using 20 runners), and your time, as not needing to write so many PRs.

Some PRs like this one, and some others are really trivial, and would be better grouped.

So, you have the choice of grouping: per issue, per folder, per your-confidence-level-on-the-fix, per file, etc.

Moreover, you're starting to get the hang of it, so your fixes should start being right on the first try more often ;)