Various Fixes and Updates

[varunagrawal]

1. Undo Python CI changes since the intended effects are not being purported. The Python-TBB CI needs more investigation to debug what the underlying issue is rather than making random guesses.
2. Fix boost collision issue for save, load, etc with boost/serialization/optional.hpp.
3. Wrapped the BarometricFactor as required in Custom Factor in MATLAB #1722.

I ran this locally on my machine Matlab and it works great!

[talregev]

It not fair that you cancel my PR, but you use my changes. Be fair.

[varunagrawal]

@talregev your PR didn't clearly state what the issue you were trying to solve was. I encountered the issue myself this week and recalled your fix.
If you want your PRs to be accepted and merged, you need to better communicate what the issue is and how you're solving it. As a reviewer, I have limited time and the onus is on you to write a good PR.

[talregev]

Even you think I wasn't communicate enough, and then you recall my fix, Do you think is a reason to steal it like that?

I will put my description here of my PR that the community will judge if it was communicate enough.

Bug fix for macos: Support new version of boost along side old version. #1721
In macos vm, recently we get a newer version of boost (1.84).
They already implemented in the new version the function that gtsam implemented.
So I am not sure the correct version of boost that it begin to happen, so I put version 1.80.
We can tune the version later (when we have another issue similar to that in different boost version).
@varunagrawal @ProfFan
Thank you for your time and effort for review this PR.

[varunagrawal]

Sounds good.

I'm re-reading your comment and it clearly doesn't mention what issue your PR is trying to fix. There is some vague mention of the MacOS VM but nothing about the fact that the latest boost version is causing a name collision with std::optional.

I looked up the error and the suggested fix was similar to yours which is why I recalled your PR. There was no ill intention and I didn't have the time to reopen your PR and go through the review back and forth, hence why I added it to this PR.

[varunagrawal]

The fix on macOS doesn't seem to generalize to Ubuntu. I'll look into it later today.

[varunagrawal]

@ProfFan FYI this PR has a better analysis of the issue that #1721 is trying to fix and thus has the better solution.

[ProfFan]

Sans one comment