CompatHelper: bump compat for OrdinaryDiffEq to 6 for package docs, (keep existing compat)

[github-actions]

This pull request changes the compat entry for the OrdinaryDiffEq package from =6.74.1 to =6.74.1, 6 for package docs.
This keeps the compat entries for earlier versions.

Note: I have not tested your package with this new compat entry.
It is your responsibility to make sure that your package tests pass before you merge this pull request.

[ChrisRackauckas-Claude]

Analysis of OrdinaryDiffEq Dependency Update

I've analyzed this PR and found important context:

Background

After this PR was created on June 19, 2024, commit 5f93405 (same day) intentionally pinned OrdinaryDiffEq to =6.74.1 with this comment in docs/make.jl:

# NOTE: OrdinaryDiffEq limited to v6.74.1 because of bug https://github.com/SciML/OrdinaryDiffEq.jl/issues/2250

Issue #2250 Status

The referenced issue was closed on June 14, 2024 as "user error" rather than being fixed. The problem was related to improper DAE initialization - initial conditions incompatible with algebraic constraints. The workaround is to use BrownFullBasicInit() for initialization.

Current Situation

• OrdinaryDiffEq v6 has continued active development with 31+ releases since 6.74.1
• Latest is v6.105.0 (December 2024)
• This PR would allow newer v6.x versions alongside the exact 6.74.1

Recommendation

Since the issue was closed as user error (not fixed), the DAE initialization problem may still exist in newer versions. However, there have been many improvements and bug fixes in the 31 releases since 6.74.1.

Options:

1. Test with latest v6: Update to latest v6.x and verify docs build successfully
2. Keep the pin: Close this PR and keep =6.74.1 until we verify newer versions work
3. Hybrid approach: Accept this PR's change (=6.74.1, 6) but test thoroughly before merging

I recommend Option 1 - update to the latest v6.x (6.105.0), test the docs build, and if successful, remove the version pin entirely. The codebase may have been fixed to handle DAE initialization properly, or the newer OrdinaryDiffEq versions may have better error messages/handling.

cc @ChrisRackauckas - Would you like me to test updating to v6.105.0 and see if the docs still build correctly?

🤖 Generated with Claude Code