[Arith] Fix dom_map var leak in IntervalSet relaxation

[guan404ming]

Why

#19927 made IntervalSetEvaluator::Eval(IntervalSet) keep the original parametric bound when relaxation loses it to infinity, even if that bound references a var being relaxed out. The leaked var reached allocation extents outside its scope, so compiling cumsum for CUDA crashed with variable i has been used before definition in CI.

How

• Keep the original bound only if it does not reference any dom_map_ var; dom_constraints_ vars stay parametric, preserving the behavior [ARITH][TIR] Track positive loop extents in analyzer visitors #19927 intended.
• Added test_relax_does_not_leak_dom_map_vars, which fails without the fix.

[gemini-code-assist]

Code Review

This pull request prevents variables being relaxed out from leaking outside their scope by ensuring that preserved parametric bounds do not reference them. A helper lambda f_depends_on_relaxed is introduced in IntervalSetEvaluator to check for these dependencies, and a corresponding unit test has been added to verify this behavior. There are no review comments, so no additional feedback is provided.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[tqchen]

i think we are moving towards a directly that over complicates things here i will do a separate followup

[tqchen]

#19963

[guan404ming]

Make sense, thanks @tqchen!