Identification of caller for ivy-read

[ThibautVerron]

Hi,

With ivy, projectile uses ivy-read for completion. This function takes a named parameter :caller that allows ivy to change its behavior (e.g. ivy-rich annotations, or display-buffer behavior). As it stands, projectile always sets this parameter to projectile-completing-read.

This PR changes it to set a different caller according to what is being read: projectile-read-project for a project name, projectile-read-buffer for a buffer name, projectile-read-file for a file name, projectile-read-directory for a directory.

Based on the changelog, the purpose appears to be similar to #1892.

It is also similar to what counsel-projectile does, but that package is much larger in scope, and more opinionated. This PR won't change user-facing behavior without additional configuration on their side.

One possible friction point is if a user has already configured eg ivy-rich to match on 'projectile-completing-read. Then they will need to adjust their configuration to the new identifiers. If that is a big problem, we can add a configuration variable to make the new behavior opt-in or opt-out.

---

Examples of what can be done (requires additional code)

Show the current branch of the projects (very useful when juggling between multiple worktrees):

Have the same information with projectile-switch-buffer as with switch-buffer:

---

Before submitting a PR make sure the following things have been done (and denote this
by checking the relevant checkboxes):

• The commits are consistent with our contribution guidelines
• You've added tests (if possible) to cover your change(s)
• All tests are passing (eldev test)
• The new code is not generating bytecode or M-x checkdoc warnings
• You've updated the changelog (if adding/changing user-visible functionality)
• You've updated the docs (when adding new project types, configuration options, commands, etc)

Thanks!