One browser test command + ability to isolate execution to individual components

[pwolfert]

Right now we have the following commands:

"test:unit:coverage"
"test:unit"
"test:unit:update"
"test:unit:preact"
"test:unit:wc"
"test:browser:all"
"test:browser"
"test:browser:smoke"
"test:browser:interaction"
"test:browser:update"
"test:browser:storybook-docs"
"test:browser:storybook-docs:update"
"test:browser:examples"
"test:browser:examples:update"

I'd like to actually consolidate test:browser:interaction into the normal browser test run and not have it use a separate Playwright config. I'd like it to be subject to the same smoke-test/full-test logic as the browser tests. The new command list would look like:

• (All the unit test commands)
• "test:browser"
• "test:browser:smoke"
• "test:browser:update"
• "test:browser:all"
• "test:browser:interaction"
• "test:storybook-docs" (rename)
• "test:storybook-docs:update" (rename)
• "test:examples" (rename)
• "test:examples:update" (rename)

We might be able to combine it with some utility to only execute tests for a particular component so that when we're updating snapshots on our branches, we can focus in on one component and not wait 30 minutes. Like you could do

yarn test:browser --component="Dropdown"
# Or
yarn test:browser:update --component="Dropdown"
# And still pass other arbitrary arguments to Playwright like
yarn test:browser --component="Dropdown" --list

In order to do that, we'd have to pre-process the command with a script, so we'd be forwarding yarn test:browser to that script.

[pwolfert]

Implemented (not exactly the same way) in #3006