Fix dangerous intarray function

[AndersAstrand]

#592 got me looking at this function and it had some weirdness to it.

This is a better version of the function I think, but maybe the pfree() is unnecessary?

[AndersAstrand]

We can of course keep using a buffer in the stack if we really think this is performance critical. It would be simple to do. I think StringInfo feels like the postgres way to do this though. I would probably prefer to analyze what context this is used in if so. Maybe we can just return the buffer if it's palloced instead of copying it at the end.

[codecov]

Codecov Report

❌ Patch coverage is 83.33333% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 85.32%. Comparing base (966bf24) to head (e3121b8).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
pg_stat_monitor.c	83.33%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #602      +/-   ##
==========================================
- Coverage   85.61%   85.32%   -0.30%     
==========================================
  Files           3        3              
  Lines        1342     1342              
  Branches      215      215              
==========================================
- Hits         1149     1145       -4     
- Misses         91       93       +2     
- Partials      102      104       +2     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.