Eval more model, control randomization & auto read webpage via workflow

- Control auto read webpage via eval workflow. Prefix env var with KHOJ_
  Default to false as it is the default that is going to be used in prod
  going forward.

- Set openai api key via input param in manual eval workflow runs
  - Simplify evaluating other chat models available over openai
    compatible api via eval workflow.
  - Mask input api key as secret in workflow.
  - Discard unnecessary null setting of env vars.

- Control randomization of samples in eval workflow.
  If randomization is turned off, it'll take the first SAMPLE_SIZE
  items from the eval dataset instead of a random collection of
  SAMPLE_SIZE items.

.github/workflows/run_evals.yml

@@ -50,11 +50,32 @@ on:
         required: false
         default: 5
         type: number
+      openai_api_key:
+        description: 'OpenAI API key'
+        required: false
+        default: ''
+        type: string
       openai_base_url:
         description: 'Base URL of OpenAI compatible API'
         required: false
         default: ''
         type: string
+      auto_read_webpage:
+        description: 'Auto read webpage on online search'
+        required: false
+        default: 'false'
+        type: choice
+        options:
+          - 'false'
+          - 'true'
+      randomize:
+        description: 'Randomize the sample of questions'
+        required: false
+        default: 'true'
+        type: choice
+        options:
+          - 'false'
+          - 'true'
 
 jobs:
   eval:
@@ -92,7 +113,14 @@ jobs:
 
       - name: Get App Version
         id: hatch
-        run: echo "version=$(pipx run hatch version)" >> $GITHUB_OUTPUT
+        run: |
+          # Mask relevant workflow inputs as secret early
+          OPENAI_API_KEY=$(jq -r '.inputs.openai_api_key' $GITHUB_EVENT_PATH)
+          echo ::add-mask::$OPENAI_API_KEY
+          echo OPENAI_API_KEY="$OPENAI_API_KEY" >> $GITHUB_ENV
+
+          # Get app version from hatch
+          echo "version=$(pipx run hatch version)" >> $GITHUB_OUTPUT
 
       - name: ⏬️ Install Dependencies
         env:
@@ -115,13 +143,13 @@ jobs:
           KHOJ_MODE: ${{ matrix.khoj_mode }}
           SAMPLE_SIZE: ${{ github.event_name == 'workflow_dispatch' && inputs.sample_size || 200 }}
           BATCH_SIZE: "20"
-          RANDOMIZE: "True"
+          RANDOMIZE: ${{ github.event_name == 'workflow_dispatch' && inputs.randomize || 'true' }}
           KHOJ_URL: "http://localhost:42110"
-          KHOJ_DEFAULT_CHAT_MODEL: ${{ github.event_name == 'workflow_dispatch' && inputs.chat_model || 'gemini-2.0-flash' }}
           KHOJ_LLM_SEED: "42"
+          KHOJ_DEFAULT_CHAT_MODEL: ${{ github.event_name == 'workflow_dispatch' && inputs.chat_model || 'gemini-2.0-flash' }}
           KHOJ_RESEARCH_ITERATIONS: ${{ github.event_name == 'workflow_dispatch' && inputs.max_research_iterations || 5 }}
+          KHOJ_AUTO_READ_WEBPAGE: ${{ github.event_name == 'workflow_dispatch' && inputs.auto_read_webpage || 'false' }}
           GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
-          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           OPENAI_BASE_URL: ${{ github.event_name == 'workflow_dispatch' && inputs.openai_base_url || '' }}
           SERPER_DEV_API_KEY: ${{ matrix.dataset != 'math500' && secrets.SERPER_DEV_API_KEY || '' }}
           OLOSTEP_API_KEY: ${{ matrix.dataset != 'math500' && secrets.OLOSTEP_API_KEY || ''}}