Ensure we don't get previously set honoree values in receipts

[jmcclelland]

Overview

When processing a series of recurring contributions, the honoree info from previously processed contributions show up in the receipts of contributions processed later.

@adixon @KarinG - We encountered this bug with a group using Iats - and it may impact Iats more then other payment processors because Iats processes recurring contributions in single run rather then via individual webhook calls.

Before

When running a series of recurring contributions, if a contribution includes an honoree (soft credit), then that honoree will show up in all subsequent receipts that do not specify an honoree.

Here's what the first (correct) receipt looks like:

And a later (incorrect) receipt might look like this:

After

By manualling clearing the variable, the incrrect data does not show up.

Technical Details

This behavior and fix are tested against 5.39. I'm sure I should submit a test to lock it in, but am a bit overwhelmed by all the moving parts of setting up the recurring contribution, checking the email etc.

[civibot]

(Standard links)

• If this is your first pull-request for CiviCRM, please browse CONTRIBUTING.md for information about the development and testing processes.
• If you are reviewing this pull-request, you may wish to consult the test sites and the Review Standards (long template, short template).

[eileenmcnaughton]

@jmcclelland yeah lots of stuff moving here at the moment - I'll try to review this & likely alter to move with the parts...

[demeritcowboy]

This seems ok to me.
Regarding a test it might be possible to do some basic setup with two regular non-recurring contributions, the first with soft-credit, and then just call static function sendEmail twice.

Minor notes:

• honorName doesn't seem to be used in the receipt, only on the confirmation/thanks page, but I see no harm setting it to null here.
• some param weirdness, passing a Smarty to formatHonoreeProfileFields() when it's expecting a form, but they happen to both have assign() functions that do the same thing.

[eileenmcnaughton]

Thanks - I definitely hadn't gotten around to doing a better fix so good to have something in