rename memories to conversations

[mdmohsin7]

part of #1973

• move memories_db to conversations_db
• replace instances of memories_db with conversations_db
• rename redis memory methods to conversations and update their instances
• rename memories specific models
• create new routes for conversations
• refactor memories utils
• test all flows in old app with new changes
• update conversation routes in app

[mdmohsin7]

Deployment Plan:

• Deploy Backend on Dev Env
• Deploy App on Dev Env and test all flows one last time
• Deploy Backend on Prod Env
• Deploy App on Prod Env

No database changes are required for this PR. The backend is backwards compatible with the app. After 1-2 weeks of releasing this version to prod, the old endpoints and models can be safely deleted permanently

@beastoin pls review

[beastoin]

lgtm @mdmohsin7

btw before merging, could you create a new PR to the development branch, merge it, then deploy it to dev env and check it ?

if everything works well, come back to me with the result then we will proceed merging on main branch.

[mdmohsin7]

lgtm @mdmohsin7

btw before merging, could you create a new PR to the development branch, merge it, then deploy it to dev env and check it ?

if everything works well, come back to me with the result then we will proceed merging on main branch.

Will do sir, and then will get back to you

[mdmohsin7]

Did not face any issues on the dev deployment. Did see the below two issues in logs but they are not related to my changes as far as I know. The below vad issue was causing problems with local sync (when I closed the app and opened again, the backend kinda got stuck in a loop or something due to this, and it wasn't working until I redeployed the backend again). This very same issue happens if I deploy to dev env from main branch.

@beastoin

[beastoin]

lgtm @mdmohsin7

vad - nvm, just because you could not connect to the VAD (check your .env > VAD HOST variable)
langsmith - we've already disabled it for a month, var LANGCHAIN_TRACING_V2 - false.