fix local bundled paths for template resources

[AjayThorve]

While #1821 was partially resolved thanks to #1830 , the template resources were still not loading due to the same path issues. This PR aims to resolve those. Testing in a jupyter_server_proxy env and jupyterlab without proxy seems to work for the changes proposed.

This allows the following example to work:

import panel as pn
import numpy as np
import holoviews as hv

pn.extension()
react = pn.template.ReactTemplate(title='React Template')

pn.config.sizing_mode = 'stretch_both'

xs = np.linspace(0, np.pi)
freq = pn.widgets.FloatSlider(name="Frequency", start=0, end=10, value=2)
phase = pn.widgets.FloatSlider(name="Phase", start=0, end=np.pi)

@pn.depends(freq=freq, phase=phase)
def sine(freq, phase):
    return hv.Curve((xs, np.sin(xs*freq+phase))).opts(
        responsive=True, min_height=400)

@pn.depends(freq=freq, phase=phase)
def cosine(freq, phase):
    return hv.Curve((xs, np.cos(xs*freq+phase))).opts(
        responsive=True, min_height=400)

react.sidebar.append(freq)
react.sidebar.append(phase)

# Unlike other templates the `ReactTemplate.main` area acts like a GridSpec 
react.main[:4, :6] = pn.Card(hv.DynamicMap(sine), title='Sine')
react.main[:4, 6:] = pn.Card(hv.DynamicMap(cosine), title='Cosine')

react.show(port=5000) # go to localhost:5000
# OR
react.show(websocket_origin="localhost:8888", port=5000) # and then go to localhost:8888/proxy/5000/

cc @philippjfr

[codecov]

Codecov Report

Merging #1891 (a0aeec4) into master (316ebbc) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #1891   +/-   ##
=======================================
  Coverage   85.10%   85.10%           
=======================================
  Files         152      152           
  Lines       18814    18815    +1     
=======================================
+ Hits        16011    16012    +1     
  Misses       2803     2803           

Impacted Files	Coverage Δ
panel/template/base.py	72.77% <100.00%> (+0.07%)	⬆️
panel/util.py	83.47% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 316ebbc...a0aeec4. Read the comment docs.

[philippjfr]

Thanks @AjayThorve. Will get a dev release out tomorrow and may release 0.10.3 with some backported fixes including this one.