feat: add specific executor param

[samsja]

Context

ATM when you pass a parameters to an Executor client.index(docs, parameters = {...} every Executor will receive all of them.
Meaning that if you want to pass different parameters to each Executor you will need to change the name.That's problematic because some Executor except the same parameters, like traversal_path.

For instance you can't embed on the chunk with one executor and embed on the root with an other one because you can only pass one traversal_path.

We need to be able to send parameters to different Executor.

original issue here : #4919
this PR take over the old proposition which we decide not to go for : #4923

Proposal for a Solution

Using the key__executor_name instead of key to precise that it should only be sent to the given Executor.

The __ operator is use for this kind of things in the python world. We use it in docarray : da[:,'tags__key'].

This is actually the exact same syntax of scikit-learn pipeline.

Parameters of the estimators in the pipeline can be accessed using the <estimator>__<parameter> syntax

from jina import Flow

f = Flow().add(name='exec1').add(name='exec2')

f.post(parameters= {'key1_exec1': val1}' # like this Exec1 will receive {'key1': val1} in their params and exec2 will receive empty.
f.post(parameters= {'key1': val1}' # like this Exec1  and Exec2 will receive {'key1': val1} 

Moreover : regarding performance issue of deserialization on the gateway side I suggest to avoid doing any deserialization on the gateway side by sending the full parameters to the Worker and only passing the one relevant to the Executor

What this PR do:

• Allow to use the __ syntax to pass executor to one Executor
• tests
• documentation

What still need to be fix/done:

• find a better way to retrieve the original name of the Executor
• better pasring machanism, something like __results__ or __something__ should not be considerate while parsing

[codecov]

Codecov Report

Merging #4934 (004a874) into master (1cd8645) will decrease coverage by 2.74%.
The diff coverage is 97.05%.

@@            Coverage Diff             @@
##           master    #4934      +/-   ##
==========================================
- Coverage   87.86%   85.12%   -2.75%     
==========================================
  Files         111      111              
  Lines        8763     8794      +31     
==========================================
- Hits         7700     7486     -214     
- Misses       1063     1308     +245     

Flag	Coverage Δ
jina	85.12% <97.05%> (-2.68%)	⬇️

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

Impacted Files	Coverage Δ
.../runtimes/request_handlers/data_request_handler.py	90.68% <97.05%> (-3.94%)	⬇️
jina/jaml/parsers/flow/v1.py	63.15% <0.00%> (-35.09%)	⬇️
jina/orchestrate/flow/base.py	71.91% <0.00%> (-18.34%)	⬇️
jina/jaml/parsers/__init__.py	75.60% <0.00%> (-17.08%)	⬇️
jina/serve/executors/__init__.py	75.41% <0.00%> (-11.18%)	⬇️
jina/serve/runtimes/gateway/http/__init__.py	90.19% <0.00%> (-7.85%)	⬇️
jina/serve/runtimes/gateway/websocket/__init__.py	90.19% <0.00%> (-7.85%)	⬇️
jina/serve/executors/decorators.py	91.66% <0.00%> (-6.25%)	⬇️
...a/orchestrate/deployments/config/docker_compose.py	94.55% <0.00%> (-5.45%)	⬇️
... and 20 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 6d7811d...004a874. Read the comment docs.

[JoanFM]

for the logic u mentioned it should be 'executorname__param` and not the other way around right?

[JoanFM]

I am not sure this solution can be handled. Since the name of the Executor w.r.t the Flow is not known by the Deployment. Imagine an External Executor cannot have the same name for every Flow

[JoanFM]

no need to copy

[JoanFM]

this is already a dict generated from protobuf