fix: Update Gemini model to non-preview version

The default Gemini model was updated from a preview version to the stable
`gemini-2.5-flash-lite`. This change was applied across Tekton pipeline
configurations and documentation for issue creation, Jira integration, and PR
CI.

Signed-off-by: Chmouel Boudjnah <chmouel@redhat.com>

Author: chmouel

.tekton/issue-create.yaml

@@ -21,7 +21,7 @@ spec:
     - name: repo_name
       value: "{{repo_name}}"
     - name: gemini_model
-      value: "gemini-2.5-flash-lite-preview-06-17"
+      value: "gemini-2.5-flash-lite"
   pipelineSpec:
     params:
       - name: repo_url

.tekton/jira.yaml

@@ -21,7 +21,7 @@ spec:
     - name: repo_name
       value: "{{repo_name}}"
     - name: gemini_model
-      value: "gemini-2.5-flash-lite-preview-06-17"
+      value: "gemini-2.5-flash-lite"
     - name: jira_endpoint
       value: "https://issues.redhat.com"
     - name: jira_project

.tekton/pr-ci.yaml

@@ -28,7 +28,7 @@ spec:
     - name: repo_name
       value: "{{repo_name}}"
     - name: gemini_model
-      value: "gemini-2.5-flash-lite-preview-06-17"
+      value: "gemini-2.5-flash-lite"
     - name: excluded_labels
       value: "good-first-issue,help-wanted,wontfix,hack"
     - name: max_labels

hack/pr-ci/README.md

@@ -61,7 +61,7 @@ The following environment variables are required:
 
 #### Optional
 
-- `GEMINI_MODEL`: Gemini model to use (default: "gemini-2.5-flash-lite-preview-06-17")
+- `GEMINI_MODEL`: Gemini model to use (default: "gemini-2.5-flash-lite")
 - `MAX_LABELS`: Maximum number of labels to apply (default: unlimited)
 - `EXCLUDED_LABELS`: Comma-separated list of labels to exclude (default: "good-first-issue,help-wanted,wontfix,hack")
 
@@ -116,7 +116,7 @@ This package is designed to work with Tekton Pipelines as Code. Two PipelineRuns
 
 ## AI Model Configuration
 
-Both pipelines use Google Gemini AI for analysis. The model can be configured via the `GEMINI_MODEL` environment variable. The current default is `"gemini-2.5-flash-lite-preview-06-17"` which provides a good balance of speed and quality for PR analysis tasks.
+Both pipelines use Google Gemini AI for analysis. The model can be configured via the `GEMINI_MODEL` environment variable. The current default is `"gemini-2.5-flash-lite"` which provides a good balance of speed and quality for PR analysis tasks.
 
 ## Development
 

hack/pr-ci/src/pr_ci/config.py

@@ -6,7 +6,7 @@
 from dataclasses import dataclass, field
 from typing import Optional
 
-DEFAULT_MODEL = "gemini-2.5-flash-lite-preview-06-17"
+DEFAULT_MODEL = "gemini-2.5-flash-lite"
 
 # JIRA Custom Fields
 JIRA_CUSTOM_FIELDS = {