Performance of elemental_data

### 🔍 Before submitting the issue

- [X] I have searched among the existing issues

### 🐞 Description of the bug

Accessing the elemental data in a huge model takes seconds.

Mainly because all data is collected at once.

### ✅ Expected behavior

Retrieving data should be much more efficient

### 📝 Steps to reproduce

Open a huge model (ask René if you don't have one)
Ensure that the model is up-to-date (model.update())
access elemental data of a ply with many elements (ply.elemental_data)

### 📃 Relevant log output

None

Start by reproducing the delay with a large model, after running model.update(), and accessing ply.elemental_data as described. Trace how elemental data is collected and identify where the full dataset is assembled. Done means retrieving elemental data is measurably more efficient on a huge model without changing the expected result.

Written by the indexing model from the issue text.