fix: Fixed high CPU usage issue related to index

[chandrasekharan-zipstack]

What

• Made sure we perform text extraction only when indexing is needed
• Removed unused packages from SDK (which was needed for LLMWhisperer previously)
• Bumped SDK to 0.15.1

Why

• We had an issue with high CPU usage due to text extraction running all the time during index - this issue affects on-prem the most

How

...

Relevant Docs

Related Issues or PRs

Dependencies Versions / Env Variables

• Removed below

 "filetype==1.2.0",
  "pdfplumber==0.10.3",
  "pytesseract==0.3.10",

Notes on Testing

• Able to generate index only once and ensured we don't re-extract everytime (checked with a debugger)
• Able to fetch response for a prompt - didn't check eval parts though

Screenshots

Checklist

I have read and understood the Contribution Guidelines.

[jaseemjaskp]

LGTM