Some Accessibility issues

[zhou-zhoukang]

Hi Developers! Thank you for your dedication to this excellent open-source app. I noticed that there are some accessibility issues.

1. The buttons shown in the two images below can only be recognized as "button" by screen readers used by visually impaired individuals. The specific function of each button cannot be conveyed. If possible, it would be very helpful to add content descriptions to these buttons.

2. The red boxes in the image below indicate all the UI elements that can be accessed by visually impaired users using screen readers. The components are quite fragmented, requiring blind users to swipe right multiple times to obtain information sequentially. If each skill could be grouped into a single accessibility unit, it would be greatly helpful. A potential approach would be using Compose's semantic APIs to group the components together, as demonstrated in Compose's official accessibility documentation. Specifically, it would be helpful to add Modifier.semantics(mergeDescendants = true) to the Row component in the SkillRow function.

3. Additionally, adding an onClickLabel to clickable controls is also a very good practice. This can be done as shown in the code example below. Taking a button as an example, before adding the label, the screen reader would announce "Double tap to activate" when focusing on the button. After adding the label, it would announce "Double tap to enter", making the button's purpose much clearer and more intuitive.

modifiler = Modifier.semantics {  
   onClick(label = "enter") { }
},

Thank you for your hard work!