[Misc] drop py 3.8 support

[felixdittrich92]

On hold - after release 0.8.0

This PR:

• drop py3.8 support (with the next release this should be fine -> https://devguide.python.org/versions/)
• tf still keeped with lower boundary at >=2.11
• fixed to <2.16.0 because keras-core (keras3) would destroy the whole TF part - needs to be updated in another PR - breaking change
• modified the references jobs a bit for speed up

TF:

Support for Python 3.8 has been removed starting with TF 2.14. The TensorFlow 2.13.1 patch release will still have Python 3.8 support.

https://github.com/tensorflow/tensorflow/releases/tag/v2.14.0

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.80%. Comparing base (8c85c36) to head (c7c1f3d).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1457      +/-   ##
==========================================
- Coverage   95.81%   95.80%   -0.02%     
==========================================
  Files         166      166              
  Lines        7648     7648              
==========================================
- Hits         7328     7327       -1     
- Misses        320      321       +1     

Flag	Coverage Δ
unittests	95.80% <100.00%> (-0.02%)	⬇️

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

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

[felixdittrich92]

@odulcy-mindee Let's talk about this next week again i don't think that it raises any impact on your side if you use a pinned doctr version.
For development it's crucical because in our CI with py3.8 we are bound to TF <= 2.13 / which makes it also not possible to work on the breaking keras changes in >= TF 2.16 :)

[felixdittrich92]

@odulcy-mindee
If your internal usage of doctr is pinned (what i thing) we could merge this also and update the github ci jobs :)
Before this isn't merged it makes less sense to work on support for TF >2.14 (>=2.16 with keras-v3) ^^