Better estimation of value for editable sliders with fixed_start / fixed_end

[hoxbro]

Before this PR, the value will always be set to 0 or (0,1) even if the value was outside the hard bounds of fixed_start / fixed_end.

[codecov]

Codecov Report

Merging #4169 (b45e51f) into master (bd3a1cf) will increase coverage by 0.01%.
The diff coverage is 96.61%.

@@            Coverage Diff             @@
##           master    #4169      +/-   ##
==========================================
+ Coverage   83.51%   83.52%   +0.01%     
==========================================
  Files         224      224              
  Lines       33388    33527     +139     
==========================================
+ Hits        27885    28005     +120     
- Misses       5503     5522      +19     

Flag	Coverage Δ
ui-tests	37.55% <33.05%> (-0.02%)	⬇️
unitexamples-tests	74.62% <96.61%> (+0.06%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
panel/widgets/slider.py	94.94% <92.59%> (-0.24%)	⬇️
panel/tests/widgets/test_slider.py	99.05% <100.00%> (+0.16%)	⬆️
panel/command/convert.py	31.42% <0.00%> (-0.93%)	⬇️
panel/io/callbacks.py	78.83% <0.00%> (-0.58%)	⬇️
panel/tests/conftest.py	87.12% <0.00%> (-0.38%)	⬇️
panel/pane/idom.py	28.16% <0.00%> (-0.20%)	⬇️
panel/tests/ui/widgets/test_tabulator.py	98.97% <0.00%> (-0.11%)	⬇️
panel/io/ipywidget.py	0.00% <0.00%> (ø)
panel/io/jupyter_server_extension.py	0.00% <0.00%> (ø)
panel/tests/util.py	87.37% <0.00%> (+0.12%)	⬆️
... and 2 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[philippjfr]

Looks like a pain, but also looks correct.