feat(#1132): Introduce async version of our API

[dcfidalgo]

Closes #1132

TODO:

• tests
• documentation
• improve docstrings
• guides
• add examples
• Log in batches. Server performance can be compromised otherwise (NOT IN THIS PR)
• Add logging traces (success and failed logs)

[dcfidalgo]

@frascuchon I did the refactoring discussed during the call. What's missing is updating the tests, documentation/docstrings, and maybe an improved behavior of the monitoring module (several retries, some automatic chunking, error messages), but maybe this can also be tackled in a separate PR.

[codecov]

Codecov Report

Merging #1391 (0bec916) into master (31c7939) will increase coverage by 0.14%.
The diff coverage is 88.70%.

@@            Coverage Diff             @@
##           master    #1391      +/-   ##
==========================================
+ Coverage   94.53%   94.67%   +0.14%     
==========================================
  Files         130      129       -1     
  Lines        5889     5898       +9     
==========================================
+ Hits         5567     5584      +17     
+ Misses        322      314       -8     

Flag	Coverage Δ
pytest	94.67% <88.70%> (+0.14%)	⬆️

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

Impacted Files	Coverage Δ
src/rubrix/__init__.py	70.58% <ø> (ø)
src/rubrix/server/commons/es_wrapper.py	87.40% <ø> (-0.75%)	⬇️
src/rubrix/monitoring/base.py	71.42% <35.71%> (-2.11%)	⬇️
src/rubrix/client/api.py	96.26% <90.56%> (+1.09%)	⬆️
src/rubrix/client/rubrix_client.py	80.35% <100.00%> (-0.24%)	⬇️
src/rubrix/client/sdk/commons/api.py	89.13% <100.00%> (+4.75%)	⬆️
src/rubrix/client/sdk/text2text/api.py	100.00% <100.00%> (ø)
src/rubrix/client/sdk/text_classification/api.py	100.00% <100.00%> (ø)
src/rubrix/client/sdk/token_classification/api.py	100.00% <100.00%> (ø)
src/rubrix/monitoring/_flair.py	86.66% <100.00%> (-0.44%)	⬇️
... and 7 more

---

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 e4d7565...0bec916. Read the comment docs.

[frascuchon]

Since I didn't find a use case just using async log_async directly, I've included a background argument in the log function that could simplify the custom data monitoring.

future= rb.log(
    name="monitor-dataset",
    records=rb_records,
    background=True, 
)

# you can always wait for the results:
future.result()

Anyway, the async log version remains visible in API in order to discover some use case using directly async function

I've also included a minimal example in monitoring guides using bentoML + spaCy.

Please, feel free to add your comments

@dvsrepo @dcfidalgo

[dcfidalgo]

I removed the Api._log method since with the new implementation there is no need for this helper method. The example in the monitor guide looks good to me, I just added a small example in the docstring of the rb.log method to simply show the effect of setting background=True. Also improved a bit this docstring.

From my side, this is good to go.

[frascuchon]

YEAH!