In Tableau dashboards, floating elements are those that can be freely moved over the dashboard canvas without being snapped into the tiled layout. The floating elements are often indicated by their standalone frames and the ability to overlap with other dashboard items. From the provided image, the state filter and the Rate of Obesity chart can be identified as floating if they can be moved independently of the tiled grid layout that constrains other elements.

To have each record alternate between grey and white, you should change the row banding size to 1 in the Format Shading pane. This setting will ensure that each row has a different shading from the one above and below it, creating an alternating pattern of grey and white rows. References: The answer can be verified through practical application within Tableau or by consulting Tableau's official learning resources, such as Tableau Help: Format Shading or Tableau Training Pass: Formatting.

To create an image of the chart that you can add to the annual report, you can use either the Export option or the Copy option from the Worksheet menu. The Export option will allow you to save the image as a file in a specified location, name, and format. The Copy option will allow you to copy the image to the clipboard and paste it into another application, such as Microsoft Word. Both options will let you choose which elements to include in the image, such as title, caption, legend, etc. Reference:
Export Views from Tableau Desktop to Another Application
Copy and Paste Views from Tableau Desktop to Another Application

The LAST() function in Tableau returns the number of rows from the current row to the last row in the partition. When you compute using 'Order Date', it will change the partitioning of the calculation. If
'Consumer' in '2018' is the last row in its partition when computed by 'Order Date', then LAST() will return 0 for that cell.
The LAST() function in Tableau is a table calculation that returns the number of rows from the current row to the last row in the partition. The value of LAST() is 0 for the last row, increases by 1 for each row above the last row, and can be negative for rows below the current row if there are such rows in the partition.
In the provided visualization, LAST() is set to compute using Table (across). Changing the compute mode to use "Order Date" will adjust the partitioning of the data upon which the LAST() function is calculated. Since
"Order Date" is likely to be a unique value per row (assuming each order has a unique date), each row becomes its own partition.
For the Consumer segment of the year 2018, if "Order Date" is unique for each row, then the last row in each partition (in this case, each individual row) will have a LAST() value of 0 because there are no other rows in the partition - it's the last row of its own partition.
Therefore, when you change the compute mode of LAST() to use "Order Date", each cell under the Consumer segment for the year 2018 will have the LAST() value of 0, because each order date creates a partition of one, making every row the last in its partition.