not able to get forecasting plot in tabset panel in flexdashboard using actionButton and eventReactive methods of shiny

I have a dataframe (in R) containing the data which looks like this:

[https://github.com/imlemarque/Data/raw/main/Data.csv][1]

I am creating a flexdashboard in R and the code is as follows:

---
title: "Forecast"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    vertical_layout: fill
    theme:
      bg: '#FFFFFF'
      fg: '#2c3e50'
      primary: '#18bc9c'
runtime: shiny

---

```{r global, include=FALSE, warning=FALSE, message=FALSE}
# libraries
library(flexdashboard)
library(readr)       
library(tidyverse)    
library(ggridges)      
library(shinydashboardPlus)
library(shiny)
library(dplyr)
library(lubridate)
library(tidyr)
library(padr)
library(shinythemes)
library(rio)
library(shinyjs)
library(forecast)

# pre-set the bw theme.
theme_set(theme_bw())
  
# convert to proper date
df$Date1 <- df$Date
df$Date <- as.Date(df$Date, format="%m/%d/%Y")

Var <- setdiff(names(select_if(df, is.numeric)), c("Date", "CovidPeriod"))


```

Forecasting {data-icon="fa-signal"}
=======================================================================

## Row {.sidebar data-width=280 data-padding=10}

```{r}
br()
selectInput("Series2", label = "Select Series:", choices = Var, multiple = FALSE, width = "98%")
checkboxInput("covid1", "Exclude covid period?", FALSE)
output$value1 <- renderText({ input$covid1 })
hr()


# Select option for confidence interval
selectInput(inputId = "conf_int", label = "Confidence Interval:",
            choice = c("99%" = "99", "95%" = "95", "80%" = "80"), multiple = FALSE, width = "98%")
hr()

sliderInput("Horiz", "Forecast Horizon:", min = 1, max = 52, value = 13)
sliderInput("HoldOut", "Holdout Period:", value = 52, min = 24, max = 104, step = 4 )
hr()


actionButton(inputId="goButton", "Start forecasting!", class = "btn-primary btn-lg")


df2 <- reactive({
        newdf <- df
        if(input$covid){ 
            newdf <- filter(df, CovidPeriod != 0)
        }
        newdf
})



```


Row {data-height=160}
-----------------------------------------------------------------------

### KPI1 {.value-box}

```{r}

renderValueBox({
valueBox(round(0.7457475, digits = 3), icon = "fa-check-square", caption = "Model Accuracy", color = "#95d7ae")
})

```

### KPI2  {.value-box}

```{r}

renderValueBox({
valueBox(round(0.34535435, digits = 3), icon = "fa-check-square", caption = "Misclassification rate", color = "primary")
})

```

### KPI3 {.value-box}

```{r}

renderValueBox({
valueBox(round(0.858656, digits = 3), icon = "fa-check-square", caption = "Recall", color = "primary")
})
```

Row {data-height=850}
-----------------------------------------------------------------------------

### 

```{r}

# Create reactive Date Column in format YYYY-MM-DD
sliderValues2 <- reactive({

  df0 <- df2() %>% mutate(Year = as.integer(year(Date)), 
                          Month = as.integer(month(Date)), 
                          Day = as.integer(day(Date))
                          )
  df0$Date2 <- strftime(paste(df0$Year, df0$Month, df0$Day, sep="-"), "%Y-%m-%d")
  df0 %>% select(Date = "Date2", input$Series2)

  })

# Render data reactively as Table
ARIMA <- eventReactive(input$goButton, {

  if (nrow(sliderValues2())==0)
    return()
  
  sliderValues2()

})


tabsetPanel(type = "tabs",
            tabPanel("ARIMA", icon = icon("area-chart"), renderTable({ARIMA()}), value=1),
            tabPanel("ARIMA2", icon = icon("area-chart"), plotOutput("ARIMA2", width="100%", height="100%"), value=1),
            id = "timeSeriesTabs")




```

My problem is that I am not able to generate forecast using Arima model to show in the ARIMA2, nor I am able to generate final table that should be shown in ARIMA tab panel.

I am getting Error:

argument is of length zero

Could someone please help in:

1. Run the fresh Arima model, on the data reactively gathered and when clicked on Go Button
2. Getting ARIMA model plots in ARIMA tab?
3. Model performance metrics in another tab (please create it)