Indexer behaviour when key does not exist should be explicit

### Type of issue

Missing information

### Description

As per [this issue](https://stackoverflow.com/questions/79265249/does-setting-a-value-using-the-indexer-count-as-an-insertion-into-an-ordereddi) on StackOverflow, I suspect the behaviour of the `OrderedDictionary`'s indexer is to `.TryInsert` when the key does not exist, therefore meaning that setting a non-existent key is equivalent to adding a new value to the end of the list, guaranteeing that it will come last when the `OrderedDictionary` is enumerated. However this is not made explicit in the documentation and therefore could currently just be an implementation detail. I think the behaviour when setting a value that doesn't currently exist via an indexer should be made explicit in the documentation (and its behaviour should be able to be relied upon long-term).

For example, adding keys such that they are at indexes 0, 1, 2, and 3, then removing the key at index 1 and then setting a value to another non-existent key should be guaranteed to result in entries at (conceptually) indexes 0, 2, 3, and (4) rather than the new entry being inserted in the empty slot, resulting in entries at indexes 0, (1), 2, and 3.

### Page URL

https://learn.microsoft.com/en-us/dotnet/api/system.collections.generic.ordereddictionary-2.item?view=net-9.0#remarks

### Content source URL

https://github.com/dotnet/dotnet-api-docs/blob/main/xml/System.Collections.Generic/OrderedDictionary`2.xml

### Document Version Independent Id

fda90b9c-d6f5-e8a1-f381-46dd197f9890

### Article author

@dotnet-bot