ARISTOTLE: get rupture geometries from rupture.json if available about oq-engine

[ptormene]

We decided to remove the option of reading the finite fault page directly because the downloadable rupture polygon files from that page are not very reliable or consistent in format. For instance, the finite fault polygon file for the 2024 Hualien earthquake does contain the rupture coordinates: https://earthquake.usgs.gov/product/finite-fault/us7000m9g4_1/us/1712690873099/shakemap_polygon.txt, but the 2023 Kahramanmaraş finite fault polygon files are empty: https://earthquake.usgs.gov/product/finite-fault/us6000jllz_1/us/1676951251912/shakemap_polygon.txt and https://earthquake.usgs.gov/product/finite-fault/us6000jlqa_1/us/1676676771092/shakemap_polygon.txt
On the other hand, the rupture.json file from the ShakeMap downloads page is reliable and consistent, and if there is a finite fault model for the event, the rupture.json file should eventually be updated to use the finite fault polygon coordinates (but this doesn’t always happen, see below for Atacama). For instance, the 2024 Hualien earthquake rupture.json file contains the same coordinates as the finite fault polygon file: https://earthquake.usgs.gov/product/shakemap/us7000m9g4/us/1717621264531/download/rupture.json. The 2023 Kahramanmaraş ShakeMap rupture.json files contain the rupture geometries which are absent in the corresponding finite fault polygon files: https://earthquake.usgs.gov/product/shakemap/us6000jllz/us/1681495642674/download/rupture.json and https://earthquake.usgs.gov/product/shakemap/us6000jlqa/us/1699129736996/download/rupture.json.
The M7.4 earthquake in Atacama, Chile last month is a case where a finite fault model is available, and the finite fault polygon file does contain the rupture coordinates: https://earthquake.usgs.gov/realtime/product/finite-fault/us7000n05d_1/us/1721493598903/shakemap_polygon.txt, but the ShakeMap rupture.json file was not updated to use the finite fault polygon coordinates and is still a point source: https://earthquake.usgs.gov/realtime/product/shakemap/us7000n05d/us/1721354540462/download/rupture.json. This means that the final ShakeMap for this event is using a point source representation of the rupture even though a finite fault model was created for the event.
Anirudh's proposal was that we should always prefer the rupture.json file. If the rupture.json file polygonal geometry, we need to convert that into an OpenQuake rupture format. The strike and dip are not needed in this case and so the text boxes on the webui must be greyed out. If the rupture.json file contains just a point source, we need to enable the strike and dip text boxes on the webui for the user to enter values, and use those values to construct the rupture polygon along with the magnitude and hypocenter.
This means that for events like the M7.4 Atacama, we would also start from a point source even though a finite fault model is available for the event. If the user wishes to use the finite fault information, they should convert it into an OpenQuake rupture.xml file and use the upload option.
If the user uploads a rupture.xml file, the strike and dip are not needed and so the text boxes on the webui must be greyed out.
That leaves the case when the rupture.json file is not available (could happen in the first few minutes or hour after the event). In this case, we need to extract the magnitude, lat, lon, depth of the event from the USGS event page. In this case, we need to enable the strike and dip text boxes on the webui for the user to enter values, and use those values to construct the rupture polygon along with the magnitude and hypocenter.