pad: stop looping over fillers when gap is complete

[gadfort]

Changes:

• Dont continue through the filler loop once gap is filled, otherwise the cells marked as overlap permitted will get added even when they are not needed.
• Correct the test to actually cause overlaps and check for them

[gemini-code-assist]

Code Review

This pull request aims to fix an issue in the filler placement logic where it would continue to add fillers even after a gap was completely filled, especially when permit_overlaps is used. The change in ICeWall.cpp stops the iteration over different filler types once the gap is filled, which is a good improvement. The test case in skywater130_overlapping_filler.tcl has been updated to correctly trigger and test this scenario.

I've found a minor remaining issue in the logic where a single extra filler can still be placed in some conditions. My review comment provides details and a suggestion for a more complete fix.

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"