Emhass

Fix
- Fix patch for issue with paths to modules and inverters database
- Fixed code formatting, or at least trying to keep a unique format

Improvement
- On this new version we now have a new method to train a regression model using Scikit-Learn methods. This is the contribution of gieljnssns. Check the dedicated section the documentation to this new feature: [https://emhass.readthedocs.io/en/latest/mlregressor.html](https://emhass.readthedocs.io/en/latest/mlregressor.html)
- Again another bunch of nice improvements by GeoDerp:
- Added Dictionary var containing EMHASS paths
- MLForcaster error suppression
- Add `freq` as runtime parameter
- Improved documentation added README buttons
- Bumping dependencies:
- Bump h5py from 3.10.0 to 3.11.0
- Bump myst-parser from 2.0.0 to 3.0.1
- Bump skforecast from 0.11.0 to 0.12.0

Fix
- Fixed bug from forecast out method related to issue 240
- Fix patch for some issues with package file paths

Improvement
- Simplified fetch urls to relatives
- Improved code for passed forecast data error handling in utils.py
- Added new tests for forecast longer than 24h by changing parameter `delta_forecast`
- Added new files for updated PV modules and inverters database for use with PVLib
- Added a new webapp to help configuring modules and inverters: [https://emhass-pvlib-database.streamlit.app/](https://emhass-pvlib-database.streamlit.app/)
- Added a new `P_to_grid_max` variable, different from the current `P_from_grid_max` option
Fix
- style.css auto format and adjusted table styling
- Changed pandas datetime rounding to nonexistent='shift_forward' to help survive DST change
- Dropped support for Python 3.9

Improvement
- Improved documentation
- Improved logging errors on missing day info
Fix
- Missing round treatment for DST survival in utils.py
- Webui large icons fix

Fix
- Fixed web_server options_json bug in standalone fix