feat: pg vector 0.8.0；perf: app pdf enhance parse (#3962)

* perf: app pdf enhance parse

* feat: pg vector 0.8.0

* update schema default

* model sort and default image

* perf: i18n

* perf: ui tip

projects/app/src/pageComponents/account/model/ModelConfigTable.tsx

@@ -563,8 +563,10 @@ const DefaultModelModal = ({
     embeddingModelList,
     ttsModelList,
     sttModelList,
-    reRankModelList
+    reRankModelList,
+    getVlmModelList
   } = useSystemStore();
+  const vlmModelList = useMemo(() => getVlmModelList(), [getVlmModelList]);
 
   // Create a copy of defaultModels for local state management
   const [defaultData, setDefaultData] = useState(defaultModels);
@@ -703,6 +705,28 @@ const DefaultModelModal = ({
             />
           </Box>
         </Box>
+        <Box>
+          <Flex mt={4} {...labelStyles} alignItems={'center'}>
+            <Box mr={0.5}>{t('account_model:vlm_model')}</Box>
+            <QuestionTip label={t('account_model:vlm_model_tip')} />
+          </Flex>
+          <Box flex={1}>
+            <AIModelSelector
+              bg="myGray.50"
+              value={defaultData.datasetImageLLM?.model}
+              list={vlmModelList.map((item) => ({
+                value: item.model,
+                label: item.name
+              }))}
+              onchange={(e) => {
+                setDefaultData((state) => ({
+                  ...state,
+                  datasetImageLLM: vlmModelList.find((item) => item.model === e)
+                }));
+              }}
+            />
+          </Box>
+        </Box>
       </ModalBody>
       <ModalFooter>
         <Button variant={'whiteBase'} mr={4} onClick={onClose}>

projects/app/src/pageComponents/dataset/detail/Import/commonProgress/DataProcess.tsx

@@ -35,21 +35,17 @@ import FormLabel from '@fastgpt/web/components/common/MyBox/FormLabel';
 import MyNumberInput from '@fastgpt/web/components/common/Input/NumberInput';
 import QuestionTip from '@fastgpt/web/components/common/MyTooltip/QuestionTip';
 import { shadowLight } from '@fastgpt/web/styles/theme';
-import AIModelSelector from '@/components/Select/AIModelSelector';
+import { DatasetPageContext } from '@/web/core/dataset/context/datasetPageContext';
+import { useToast } from '@fastgpt/web/hooks/useToast';
 
 function DataProcess() {
   const { t } = useTranslation();
   const { feConfigs } = useSystemStore();
+  const { toast } = useToast();
 
-  const {
-    goToNext,
-    processParamsForm,
-    chunkSizeField,
-    minChunkSize,
-    maxChunkSize,
-    priceTip,
-    chunkSize
-  } = useContextSelector(DatasetImportContext, (v) => v);
+  const { goToNext, processParamsForm, chunkSizeField, minChunkSize, maxChunkSize } =
+    useContextSelector(DatasetImportContext, (v) => v);
+  const datasetDetail = useContextSelector(DatasetPageContext, (v) => v.datasetDetail);
   const { getValues, setValue, register, watch } = processParamsForm;
   const trainingType = watch('trainingType');
   const chunkSettingMode = watch('chunkSettingMode');
@@ -177,9 +173,16 @@ function DataProcess() {
                         <QuestionTip label={t('dataset:auto_indexes_tips')} />
                       </HStack>
                       <HStack flex={'1'} spacing={1}>
-                        <Checkbox {...register('imageIndex')}>
-                          <FormLabel>{t('dataset:image_auto_parse')}</FormLabel>
-                        </Checkbox>
+                        <MyTooltip
+                          label={!datasetDetail?.vlmModel ? t('common:error_vlm_not_config') : ''}
+                        >
+                          <Checkbox
+                            isDisabled={!datasetDetail?.vlmModel}
+                            {...register('imageIndex')}
+                          >
+                            <FormLabel>{t('dataset:image_auto_parse')}</FormLabel>
+                          </Checkbox>
+                        </MyTooltip>
                         <QuestionTip label={t('dataset:image_auto_parse_tips')} />
                       </HStack>
                     </HStack>

projects/app/src/pageComponents/dataset/detail/Info/index.tsx

@@ -37,7 +37,7 @@ const Info = ({ datasetId }: { datasetId: string }) => {
   const { t } = useTranslation();
   const { datasetDetail, loadDatasetDetail, updateDataset, rebuildingCount, trainingCount } =
     useContextSelector(DatasetPageContext, (v) => v);
-  const { feConfigs, datasetModelList, embeddingModelList, getVllmModelList } = useSystemStore();
+  const { feConfigs, datasetModelList, embeddingModelList, getVlmModelList } = useSystemStore();
 
   const [editedDataset, setEditedDataset] = useState<EditResourceInfoFormType>();
   const [editedAPIDataset, setEditedAPIDataset] = useState<EditAPIDatasetInfoFormType>();
@@ -52,7 +52,7 @@ const Info = ({ datasetId }: { datasetId: string }) => {
   const vectorModel = watch('vectorModel');
   const agentModel = watch('agentModel');
 
-  const vllmModelList = useMemo(() => getVllmModelList(), [getVllmModelList]);
+  const vllmModelList = useMemo(() => getVlmModelList(), [getVlmModelList]);
   const vlmModel = watch('vlmModel');
 
   const { ConfirmModal: ConfirmDelModal } = useConfirm({

projects/app/src/pageComponents/dataset/list/CreateModal.tsx

@@ -40,7 +40,7 @@ const CreateModal = ({
 }) => {
   const { t } = useTranslation();
   const router = useRouter();
-  const { feConfigs, defaultModels, embeddingModelList, datasetModelList, getVllmModelList } =
+  const { feConfigs, defaultModels, embeddingModelList, datasetModelList, getVlmModelList } =
     useSystemStore();
   const { isPc } = useSystem();
 
@@ -71,7 +71,7 @@ const CreateModal = ({
 
   const filterNotHiddenVectorModelList = embeddingModelList.filter((item) => !item.hidden);
 
-  const vllmModelList = useMemo(() => getVllmModelList(), [getVllmModelList]);
+  const vllmModelList = useMemo(() => getVlmModelList(), [getVlmModelList]);
 
   const form = useForm<CreateDatasetParams>({
     defaultValues: {