Add E2B as an optional code sandbox provider

- Specify E2B api key and template to use via env variables
- Try load, use e2b library when E2B api key set
- Fallback to try use terrarium sandbox otherwise
- Enable more python packages in e2b sandbox like rdkit via custom e2b template

- Use Async E2B Sandbox
- Parallelize file IO with sandbox
- Add documentation on how to enable E2B as code sandbox instead of Terrarium

.github/workflows/run_evals.yml

@@ -32,6 +32,14 @@ on:
         required: false
         default: 200
         type: number
+      sandbox:
+        description: 'Code sandbox to use'
+        required: false
+        default: 'terrarium'
+        type: choice
+        options:
+          - terrarium
+          - e2b
 
 jobs:
   eval:
@@ -100,6 +108,8 @@ jobs:
           SERPER_DEV_API_KEY: ${{ matrix.dataset != 'math500' && secrets.SERPER_DEV_API_KEY }}
           OLOSTEP_API_KEY: ${{ matrix.dataset != 'math500' && secrets.OLOSTEP_API_KEY }}
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
+          E2B_API_KEY: ${{ inputs.sandbox == 'e2b' && secrets.E2B_API_KEY }}
+          E2B_TEMPLATE: ${{ vars.E2B_TEMPLATE }}
           KHOJ_ADMIN_EMAIL: khoj
           KHOJ_ADMIN_PASSWORD: khoj
           POSTGRES_HOST: localhost
@@ -148,6 +158,7 @@ jobs:
           echo "**$(head -n 1 *_evaluation_summary_*.txt)**" >> $GITHUB_STEP_SUMMARY
           echo "- Khoj Version: ${{ steps.hatch.outputs.version }}" >> $GITHUB_STEP_SUMMARY
           echo "- Chat Model: Gemini 2.0 Flash" >> $GITHUB_STEP_SUMMARY
+          echo "- Code Sandbox: ${{ inputs.sandbox}}" >> $GITHUB_STEP_SUMMARY
           echo "\`\`\`" >> $GITHUB_STEP_SUMMARY
           tail -n +2 *_evaluation_summary_*.txt >> $GITHUB_STEP_SUMMARY
           echo "" >> $GITHUB_STEP_SUMMARY