Fix Xiaomi H1 alternative signature

[dumpfheimer]

Proposed change

It seems like this commit kicked a device signature for Xiaomi H1 switches: 4f5abdd
This PR adds it back.

Additional information

Checklist

• The changes are tested and work correctly
• pre-commit checks pass / the code has been formatted using Black
• Tests have been added to verify that the new code works

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (aeca76d) 87.35% compared to head (b8d7a49) 87.35%.

Additional details and impacted files

@@           Coverage Diff           @@
##              dev    #2806   +/-   ##
=======================================
  Coverage   87.35%   87.35%           
=======================================
  Files         287      287           
  Lines        8835     8837    +2     
=======================================
+ Hits         7718     7720    +2     
  Misses       1117     1117           

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

[TheJulianJES]

Hmm, unless I'm missing something, I don't see where that quirk was removed. Maybe it was using a quirk without MODELS_INFO before and the MODELS_INFO is missing lumi.switch.n2aeu1 now.
Do you still know which (exact) quirk it was using before? (EDIT: Likely XiaomiOpple2ButtonSwitchFace1)

I'll have a look.

Please also attach device diagnostics and/or the device signature. It can be downloaded from the device page.
(Without the quirk applied!)

[TheJulianJES]

Looks like it matches the XiaomiOpple2ButtonSwitchFace1 signature too:

zha-device-handlers/zhaquirks/xiaomi/aqara/opple_switch.py

Lines 277 to 281 in aeca76d

	class XiaomiOpple2ButtonSwitchFace1(XiaomiOpple2ButtonSwitchBase):
	"""Xiaomi Opple 2 Button Switch. Face 1."""
	signature = {
	MODELS_INFO: [(LUMI, "lumi.switch.b2naus01")],

So, (LUMI, "lumi.switch.n2aeu1") just needs to be added to the MODELS_INFO there.
Maybe we should move it to the "H1 double" file too then? Although the lumi.switch.b2naus01 isn't a H1 switch from what I can see.

[TheJulianJES]

Okay, it seems like (at least the US models of) Aqara switches sometimes report a different signature when paired. This might explain some things.
In quirks, we need to match all signatures then.

I'll fix #2824 first and then see how many signatures also match the H1 models.

[TheJulianJES]

Superseded by #2826
I'll see if we can get that into a patch release of HA.

Thanks for making me aware btw!

[dumpfheimer]

Great, thank you!

In addition to this, I made some configuration entities for the H1 and was wondering if there was a better way to match all of the devices using a similar configuration, to prevent a HA PR every time a device is added:
home-assistant/core@a698090

right now the matching works by using a quirk ID, eg

@CONFIG_DIAGNOSTIC_MATCH(
    cluster_handler_names="opple_cluster", quirk_ids="zhaquirks.xiaomi.aqara.opple_switch.XiaomiOpple2ButtonSwitchFace1"
)

Could you point me in the right direction for how I could match these configuration entities to catch all devices?