feat: freeze jina version in hub build

[JoanFM]

Changes introduced
Freeze jina version to the core version that builds the hub image

resolved #1378

[github-actions]

Latency summary

Current PR yields:

• 😶 index QPS at 1695, delta to last 3 avg.: +0%
• 😶 query QPS at 31, delta to last 3 avg.: -3%

Breakdown

Version	Index QPS	Query QPS
current	1695	31
0.9.1	1699	32
0.8.22	1703	31
0.8.21	1700	32

Backed by latency-tracking. Further commits will update this comment.

[nan-wang]

LGTM👍 Maybe we can add a unit test?

[JoanFM]

@nan-wang does not this conflict with ur PR in action builder?

[nan-wang]

@nan-wang does not this conflict with ur PR in action builder?

Nope. This is consistent with current behaviors. The hub-builder calls hubio for building the image and therefore the jina_version depends on which base image is used.

However, your concern is valid. The Jina versions used in hub-updater and hub-builder are implicitly coupled.

[codecov]

Codecov Report

Merging #1587 (3b27e6b) into master (35139e9) will increase coverage by 1.36%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1587      +/-   ##
==========================================
+ Coverage   83.44%   84.80%   +1.36%     
==========================================
  Files         128      128              
  Lines        6646     6721      +75     
==========================================
+ Hits         5546     5700     +154     
+ Misses       1100     1021      -79     

Impacted Files	Coverage Δ
jina/docker/checker.py	82.14% <ø> (ø)
jina/docker/hubio.py	73.10% <100.00%> (+11.46%)	⬆️
jina/logging/profile.py	69.84% <0.00%> (-0.56%)	⬇️
jina/drivers/craft.py	100.00% <0.00%> (ø)
jina/types/ndarray/generic.py	100.00% <0.00%> (ø)
jina/drivers/encode.py	94.91% <0.00%> (+0.08%)	⬆️
jina/enums.py	96.59% <0.00%> (+0.09%)	⬆️
jina/jaml/__init__.py	95.93% <0.00%> (+0.09%)	⬆️
jina/clients/helper.py	94.00% <0.00%> (+0.12%)	⬆️
jina/peapods/runtimes/asyncio/rest/__init__.py	93.18% <0.00%> (+0.15%)	⬆️
... and 17 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 35139e9...3b27e6b. Read the comment docs.

[nan-wang]

LGTM👍