fix(formatter): Stable member chain printing

[MichaReiser]

This PR simplifies the member printing logic to make sure its output is stable.
Interestingly, this is slightly improving the prettier compatibility

Before:
**File Based Average Prettier Similarity**: 69.29%  
**Line Based Average Prettier Similarity**: 64.28%  

After:
**File Based Average Prettier Similarity**: 69.39%  
**Line Based Average Prettier Similarity**: 64.40%  

But unblocks my work on refactoring the printer behaviour (which seems to trigger this now a lot), and #2458 . The work on improving the formatting is tracked in #2421

Test Plan

I rebased #2458 (need to find a way to push this change) and the tests now pass.

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Pages

Latest commit:	5e8b0a2
Status:	✅  Deploy successful!
Preview URL:	https://e4a1592b.tools-8rn.pages.dev

View logs

[github-actions]

Parser conformance results on ubuntu-latest

js/262

Test result	main count	This PR count	Difference
Total	45878	45878	0
Passed	44938	44938	0
Failed	940	940	0
Panics	0	0	0
Coverage	97.95%	97.95%	0.00%

jsx/babel

Test result	main count	This PR count	Difference
Total	39	39	0
Passed	36	36	0
Failed	3	3	0
Panics	0	0	0
Coverage	92.31%	92.31%	0.00%

ts/babel

Test result	main count	This PR count	Difference
Total	588	588	0
Passed	519	519	0
Failed	69	69	0
Panics	0	0	0
Coverage	88.27%	88.27%	0.00%

ts/microsoft

Test result	main count	This PR count	Difference
Total	16257	16257	0
Passed	12391	12391	0
Failed	3866	3866	0
Panics	0	0	0
Coverage	76.22%	76.22%	0.00%

[github-actions]

Playground for commit e0d0668

[leops]

I does result in quite a few regressions but I guess that's fine as the intent is to go back to the chain printing later on and fully align it with Prettier when we have the appropriate IR to do so

[leops]

This method is now doing the same thing as into_format_elements, is there any point in keeping it ?

[MichaReiser]

I reverted this change with the added benefit that there are now fewer regressions (overall even an increase in prettier compatibility)

[MichaReiser]

@cpojer You should be able to land your PR after this :) I think it's improving our new metric by about 1%!

[MichaReiser]

I does result in quite a few regressions but I guess that's fine as the intent is to go back to the chain printing later on and fully align it with Prettier when we have the appropriate IR to do so

Yeah, unfortunately. But it's sometimes necessary to go a few steps back to be able to move forward again. I think this is one such situation (I still have to fight with some other unstable results but at least fewer of them)

[ematipico]

I am not sure if this was the correct way to go. I hope it is. It seems that lot of logic taken from prettier was removed.

[MichaReiser]

I agree that this hasn't been ideal but it was, in my view necessary, to make progress on changing the underlying printer architecture without being stuck for days (It still took me 1.5 weeks). It should be easier to re-add this logic now, that the printer works the same as prettier.

Something I noticed is that we often form groups in the exact opposite order than prettier. For example, prettier groups left && where we group && right. These are things that we need to re-visit to align our formatting.