docs: clarify atomic_dipole meaning for DPLR models (#4979)

✅ **UPDATED: Fixed Documentation Structure and Language**

## Changes Made Based on Review Comments

### 📝 Addressed Feedback from @njzjz

**Issue**: The inserted section broke the original content structure and
the language was too informal.

**Changes**:

1. **Fixed broken documentation structure**
- Removed the standalone section that was inserted in the middle of
"Train a deep Wannier model for Wannier centroids"
- Integrated the atomic_dipole explanation into the existing flow
without breaking the section

2. **Formalized language and presentation**
- Converted informal bullet points and step-by-step lists to more formal
prose
- Integrated the explanation naturally into the existing sentence about
atomic_dipole being trained as labels
- Maintained technical accuracy while using more formal documentation
style

3. **Restored proper content flow**
- The section now flows properly from the loss configuration to the
atomic dipole explanation to the training commands
- No longer interrupts the original structure of the "Train a deep
Wannier model" section

### 📖 Final Result

The DPLR documentation now includes a concise, formal explanation of
what atomic_dipole means in the DPLR context (displacement from atom to
Wannier centroid) integrated seamlessly into the existing training
workflow description, without breaking the document structure.

Fixes #4970.

<!-- START COPILOT CODING AGENT TIPS -->
---

💬 Share your feedback on Copilot coding agent for the chance to win a
$200 gift card! Click
[here](https://survey3.medallia.com/?EAHeSx-AP01bZqG0Ld9QLQ) to start
the survey.

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: njzjz <9496702+njzjz@users.noreply.github.com>

Author: Copilot

doc/model/dplr.md

@@ -69,7 +69,7 @@ The loss section is provided as follows
     },
 ```
 
-so that the atomic dipole is trained as labels. Note that the NumPy compressed file `atomic_dipole.npy` should be provided in each dataset.
+so that the atomic dipole is trained as labels. Note that the NumPy compressed file `atomic_dipole.npy` should be provided in each dataset. In the context of DPLR models, the atomic dipole data represents the displacement vector from each atom to its associated Wannier centroid (WC), which can be calculated as `atomic_dipole = wannier_centroid_position - atom_position` from DFT calculations using tools such as VASP with Wannier90.
 
 The training and freezing can be started from the example directory by