Introduce MongoDB sharding rules to Documents and Document-wide collections

[sejongk]

What this PR does / why we need it:
This PR introduces MongoDB sharding rules to documents and document-wide collections (e.g. Changes, SyncedSync, Snapshots).

Which issue(s) this PR fixes:

Addresses #673

Special notes for your reviewer:

Does this PR introduce a user-facing change?:

Additional documentation:

Checklist:

• Added relevant tests or not required
• Didn't break anything

[codecov]

Codecov Report

Attention: 183 lines in your changes are missing coverage. Please review.

Comparison is base (eaefcdc) 50.17% compared to head (46f7010) 50.87%.
Report is 1 commits behind head on main.

Files	Patch %	Lines
server/backend/database/mongo/client.go	64.70%	48 Missing ⚠️
server/backend/database/memory/database.go	52.63%	26 Missing and 1 partial ⚠️
server/backend/database/mongo/registry.go	65.78%	18 Missing and 8 partials ⚠️
server/rpc/yorkie_server.go	69.23%	16 Missing ⚠️
server/backend/database/client_info.go	71.73%	13 Missing ⚠️
server/backend/sync/memory/pubsub.go	45.83%	13 Missing ⚠️
server/rpc/admin_server.go	52.17%	11 Missing ⚠️
server/backend/database/snapshot_info.go	0.00%	9 Missing ⚠️
client/client.go	0.00%	8 Missing ⚠️
api/types/resource_ref_key.go	57.14%	6 Missing ⚠️
... and 2 more

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #734      +/-   ##
==========================================
+ Coverage   50.17%   50.87%   +0.70%     
==========================================
  Files          69       70       +1     
  Lines       10165    10168       +3     
==========================================
+ Hits         5100     5173      +73     
+ Misses       4526     4473      -53     
+ Partials      539      522      -17     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[hackerwins]

Thanks for your contribution.

I looked at the code up to the memory DB. and I have a question.

MemoryDB runs standalone, and it is not sharded based on shard keys. If we receive refKey due to change in interface, we only need to utilize the id. So there's no necessity to use the key in the implementation. What are your thoughts on this?

[sejongk]

Thanks for your contribution.

I looked at the code up to the memory DB. and I have a question.

MemoryDB runs standalone, and it is not sharded based on shard keys. If we receive refKey due to change in interface, we only need to utilize the id. So there's no necessity to use the key in the implementation. What are your thoughts on this?

I think that's also a good idea.
Previously, I heard that the MemoryDB is utilized for the testing purpose, so I tried to diminish the gap between the implementation of MongoDB and MemoryDB for the testing consistency.

[hackerwins]

LGTM. 👍