fix: post should handle version properly

[bwanglzu]

Context

Not a big fan of the solution, but the current implementation of post is not usable:

1. versions are not handled. e.g. executor versions such as v0.9.1-gpu is commonly used while using an executor. The current implementation only allow us to use default, which is latest. This result in host parsing not correctly parse the executor versions.
2. keyword arguments are not parsed into Executor, such as uses_with, volumes.

This could be better handled by unifying jina.hubble.helper.parse_hub_uri, or even better to create an extra on argument to the post method, but it will introduce breaking change, so i did not implement it.

It is already working. But i would do a bit more refactoring, move the logic of urlparsing out of post, to make it unit testable.

Why this is needed?

from docarray import DocumentArray, Document

da = DocumentArray([Document(text='hello')])
da = da.post(
    'jinahub+docker://FinetunerExecutor/v0.9.2/encode',
    uses_with={'artifact': '/mnt/finetune-quora-19.zip'},
    volumes=['/Users/bo/Documents/work2/test/my-model/:/mnt'],
)
assert da[0].embedding.shape==(768,)

[codecov]

Codecov Report

Merging #441 (07fef96) into main (25a59af) will increase coverage by 1.69%.
The diff coverage is 96.77%.

@@            Coverage Diff             @@
##             main     #441      +/-   ##
==========================================
+ Coverage   84.93%   86.62%   +1.69%     
==========================================
  Files         134      134              
  Lines        6466     6477      +11     
==========================================
+ Hits         5492     5611     +119     
+ Misses        974      866     -108     

Flag	Coverage Δ
docarray	86.62% <96.77%> (+1.69%)	⬆️

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

Impacted Files	Coverage Δ
docarray/array/mixins/post.py	93.02% <96.77%> (+5.52%)	⬆️
docarray/array/mixins/find.py	87.35% <0.00%> (+1.14%)	⬆️
docarray/helper.py	82.80% <0.00%> (+1.35%)	⬆️
docarray/array/storage/sqlite/seqlike.py	84.31% <0.00%> (+1.96%)	⬆️
docarray/array/queryset/lookup.py	78.10% <0.00%> (+2.18%)	⬆️
docarray/array/mixins/io/binary.py	97.45% <0.00%> (+2.54%)	⬆️
docarray/array/storage/weaviate/seqlike.py	77.41% <0.00%> (+3.22%)	⬆️
docarray/array/mixins/io/from_gen.py	83.63% <0.00%> (+3.63%)	⬆️
docarray/array/storage/weaviate/backend.py	87.50% <0.00%> (+3.67%)	⬆️
docarray/array/storage/elastic/backend.py	95.04% <0.00%> (+4.13%)	⬆️
... and 9 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 25a59af...07fef96. Read the comment docs.

[JoanFM]

Why not instead of pattern matching version just check if there are more / than expected, and u just extract the last part as the endpoint, and everything before as the host where u apply some extra logic?

[bwanglzu]

@JoanFM do you think check extra / is even more fragile?

[JoanFM]

@JoanFM do you think check extra / is even more fragile?

it is much easier to mantain

[JoanFM]

LGTM, please make sure it is clear in documentation