Coordinate system wraps OSMTILE at -180, creating multiple worlds
Here's an example. At low zoom levels, it's not desirable to repeat the world and wrap coordinates. It looks bad and it's weird (not consistent / explicable by TCRS):

This can be controlled by the Leaflet noWrap option in tiled layers (e.g. M.TemplatedTileLayer), but in WMS based layers, which are custom implemented via the M.TemplatedImageLayer (and probably also a problem in M.TemplatedFeatureLayer s, but I didn't test that), the suggested method for controlling CRS coordinate wrapping in Leaflet doesn't work, so it's probably not a Leaflet problem.
This actually was a setting in GeoServer for WMS requests:

So, the fix can be achieved by adding noWrap to M.TemplatedTileLayer options.
Applying the suggested fix above yields unsatisfactory results in some tiled layers:

We've seen this issue before at the edges of tile caches. The problem is likely rounding up instead of truncating, generating a 'true' value for tiles that are barely within bounds, which should be treated as not in bounds. But you don't know if there's going to be a response from the cache or not, so opting to be generous can lead to maps like the above, while opting to be stingy yields maps with missing tiles.
This MAY point to the need for a declarative way to tell the map engine to wrap the coordinate system TBD, based on discussion and evidence.
Another method of at least mitigating situations like the above might be to verify if a tile even mathematically exists within the TCRS bounds before deciding to request it.
Applying the suggested fix above yields unsatisfactory results in some tiled layers
I was looking into this the other day out of curiosity, I don't quite understand all of https://github.com/Maps4HTML/Web-Map-Custom-Element/issues/479#issuecomment-877468181, but according to some Leaflet issue you should set the bounds option such that tiles out of bounds aren't requested (and shouldn't yield the unsatisfactory result as described above). Maybe this is obvious to you, in that case you can disregard this comment.