doc(base): update lying doc comment of compute_display_name

[bnjbvr]

It claimed that it would immediately return when the cached display name value was computed, but that's absolutely not the case.

Spotted while reviewing a PR updating iamb to the latest version of the SDK.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 85.04%. Comparing base (1480fad) to head (248f5e7).
Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4465      +/-   ##
==========================================
- Coverage   85.06%   85.04%   -0.02%     
==========================================
  Files         283      283              
  Lines       31748    31766      +18     
==========================================
+ Hits        27005    27016      +11     
- Misses       4743     4750       +7     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[poljar]

A bit tangentially related, but why don't we have a method that does what the docs claim to do?

A method like this:

pub async fn foo() -> StoreResult<RoomDisplayName> {
    if let Some(result) = self.cached_display_name() {
        result
    } else {
        self.compute_dislpay_name().await
    }
}

We may even call this method just display_name() and people wouldn't need to think about the other two methods, unless they really care about never hitting the slow path.

[bnjbvr]

A bit tangentially related, but why don't we have a method that does what the docs claim to do?

Ah nice, I thought that the reason was that compute_display_name was defined in the base crate, and used in the main crate, but that's not true! So we could have compute_display_name be private in the base crate, and expose a public method like that you propose. Will do as a followup!