fix: better memory usage from 800+ to 500+

[yihong0618]

Summary

with this patch

without this patch

how to trace the memory

1. use https://github.com/bloomberg/memray
2. run memray run app.py and memray run app.py --debug
3. read the flamegraph found some big memory are from some package like google and jieba
4. lazy import them to better the memory usage and short the start time

flamegraph without this patch

flamegraph with this patch

Tip

Close issue syntax: Fixes #<issue number> or Resolves #<issue number>, see documentation for more details.

Screenshots

Before	After
...	...

Checklist

Important

Please review the checklist below before submitting your pull request.

• This change requires a documentation update, included: Dify Document
• I understand that this PR may be closed in case there was no previous discussion or issues. (This doesn't apply to typos!)
• I've added a test for each change that was introduced, and I tried as much as possible to make a single atomic change.
• I've updated the documentation accordingly.
• I ran dev/reformat(backend) and cd web && npx lint-staged(frontend) to appease the lint gods

[laipz8200]

LGTM, does this PR ready to merge?

[yihong0618]

My side is done maybe some other memory deduce can be a separate PR

[laipz8200]

Great! Thank you!