feat: YataiClient context to communicate with multiple Yatai instances

[ssheng]

What does this PR address?

The .yatai.yaml configuration under BENTOML_HOME already supports contexts to communicate with multiple Yatai instances through different API tokens today. However, the same functionality was not exposed to the CLI commands and YataiClient API. This PR exposes setting and selecting contexts in the CLI.

Yatai Log-in

> bentoml yatai login --api-token xxx --endpoint https://yatai1 --context yatai1
Successfully logged in as user "ssheng" in organization "yatai1".
> bentoml yatai login --api-token xxx --endpoint https://yatai2 --context yatai2
Successfully logged in as user "ssheng" in organization "yatai2".

Bento push & pull

> bentoml push iris_classifier:x67srpdqs25aicvj --context yatai1
> bentoml push iris_classifier:x67srpdqs25aicvj --context yatai2

Model push & pull

> bentoml models push iris_clf:7r5r5ftamc3dccvj --context=yatai1
> bentoml models push iris_clf:7r5r5ftamc3dccvj --context=yatai2

YataiClient API

yatai_client.push_bento(bento, context="yatai1")
yatai_client.pull_bento(bento_tag, context="yatai1")
yatai_client.push_model(model, context="yatai1")
yatai_client.pull_model(model_tag, context="yatai1")

Before submitting:

• Does the Pull Request follow Conventional Commits specification naming? Here are GitHub's
  guide on how to create a pull request.
• Does the code follow BentoML's code style, both make format and make lint script have passed (instructions)?
• Did you read through contribution guidelines and follow development guidelines?
• Did your changes require updates to the documentation? Have you updated
  those accordingly? Here are documentation guidelines and tips on writting docs.
• Did you write tests to cover your changes?

[codecov]

Codecov Report

Merging #3448 (75070f0) into main (1d91d8c) will decrease coverage by 0.36%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main    #3448      +/-   ##
==========================================
- Coverage   33.10%   32.75%   -0.36%     
==========================================
  Files         132      134       +2     
  Lines       10689    10836     +147     
  Branches     1778     1712      -66     
==========================================
+ Hits         3539     3549      +10     
- Misses       6910     7049     +139     
+ Partials      240      238       -2     

Impacted Files	Coverage Δ
src/bentoml/_internal/io_descriptors/pandas.py	39.03% <0.00%> (-3.92%)	⬇️
src/bentoml/_internal/io_descriptors/numpy.py	42.57% <0.00%> (-3.80%)	⬇️
src/bentoml/_internal/container/__init__.py	76.00% <0.00%> (-2.00%)	⬇️
src/bentoml/_internal/container/base.py	61.68% <0.00%> (-1.87%)	⬇️
src/bentoml/_internal/configuration/__init__.py	60.52% <0.00%> (-1.32%)	⬇️
src/bentoml/_internal/utils/__init__.py	44.96% <0.00%> (-0.78%)	⬇️
src/bentoml/__init__.py	100.00% <0.00%> (ø)
src/bentoml/_internal/models/model.py	77.88% <0.00%> (ø)
src/bentoml/_internal/server/server.py	0.00% <0.00%> (ø)
src/bentoml/_internal/utils/benchmark.py	0.00% <0.00%> (ø)
... and 7 more

[aarnphm]

LGTM.

[aarnphm]

I'm thinking maybe for push and pull we can be more explicit

bentoml models push ... --yatai-context

whereas for bentoml yatai --context works?

[ssheng]

I'm thinking maybe for push and pull we can be more explicit

bentoml models push ... --yatai-context

whereas for bentoml yatai --context works?

Make sense but in the case of BentoCloud, I worry that the yatai prefix maybe irrelevant or misleading.

[aarnphm]

Make sense but in the case of BentoCloud, I worry that the yatai prefix maybe irrelevant or misleading.

How about bentoml models push --remote-context?

[aarnphm]

As discussed.