Refactor faceted track selector to use more MST state

[cmdcolin]

The current faceted track selector uses mostly react hooks (and quite a few of them, useMemo, useReducer, useState, useEffect, prop drilling the useReducer/useState callbacks to multiple components, etc.) to try to synchronize relatively complex state. This changes it to mobx-state-tree which has a number of simplifying effects IMO

[codecov]

Codecov Report

Attention: 61 lines in your changes are missing coverage. Please review.

Comparison is base (05e1776) 63.24% compared to head (7fdd2e7) 63.35%.
Report is 1 commits behind head on main.

❗ Current head 7fdd2e7 differs from pull request most recent head 69fce71. Consider uploading reports for the commit 69fce71 to get more accurate results

Files	Patch %	Lines
...rc/HierarchicalTrackSelectorWidget/facetedModel.ts	64.93%	27 Missing ⚠️
...ectorWidget/components/faceted/FacetedSelector.tsx	0.00%	11 Missing ⚠️
...SelectorWidget/components/faceted/FacetFilters.tsx	0.00%	9 Missing ⚠️
...electorWidget/components/faceted/FacetedHeader.tsx	0.00%	8 Missing ⚠️
...kSelectorWidget/components/faceted/FacetFilter.tsx	0.00%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4094      +/-   ##
==========================================
+ Coverage   63.24%   63.35%   +0.10%     
==========================================
  Files        1045     1046       +1     
  Lines       30616    30653      +37     
  Branches     7314     7323       +9     
==========================================
+ Hits        19362    19419      +57     
+ Misses      11084    11062      -22     
- Partials      170      172       +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[cmdcolin]

I think this PR is still a good approach but there is a small bug where if you hide a column, it then resets the widths for all columns after that. hard to fix properly, so will put this PR in draft, and make a new one...

[cmdcolin]

actually, interestingly, the current main branch has the same bug (it resets the widths after you hide a column) so this is not necessarily a regression on this branch....still would be good to fix but this branch could go forward