Replace Selenium with Direct HTTP Requests

**Description:**

The project currently uses Selenium WebDriver to fetch data from URLs such as:
https://sisu-api.sisu.mec.gov.br/api/v1/oferta/instituicoes/uf
https://sisu-api.sisu.mec.gov.br/api/v1/oferta/instituicao/{iesCode}
[https://sisu-api.sisu.mec.gov.br/api/v1/oferta/{offerCode}/modalidades](https://sisu-api.sisu.mec.gov.br/api/v1/oferta/%7BofferCode%7D/modalidades)

These endpoints return **raw JSON**, so using Selenium to load a browser, parse HTML, and extract the JSON from a `

` tag is unnecessary. This makes the program slower, more fragile, and more resource-heavy.  

This issue proposes replacing all Selenium-based HTTP fetching with direct `HttpClient` GET requests.


---

### **Tasks**

- [ ] Create a static `HttpClient` instance, e.g.:  

  `private static readonly HttpClient client = new HttpClient();`

- [ ] Replace all `driver.Navigate().GoToUrl(url)` calls with:  

  `var json = await client.GetStringAsync(url);`

- [ ] Remove HTML parsing using HtmlAgilityPack

- [ ] Deserialize JSON directly using `JsonConvert.DeserializeObject`

- [ ] Preserve existing data structures (`Institution`, `Course`, `CourseWeights`, etc.)

- [ ] Remove Selenium and HtmlAgilityPack from the project dependencies

- [ ] Update README to reflect that Selenium is no longer required

- [ ] Ensure all current features behave identically after the change

---

### **Why This Is Good for Beginners**

- Involves simple and clear modifications  

- Improves code performance and maintainability  

- No advanced browser automation required  

- Easy to test because the APIs are public and return consistent JSON  

- Gives contributors experience with networking and JSON parsing

---

### **Optional Enhancements**

- [ ] Add error handling for network failures  

- [ ] Implement timeout for requests  

- [ ] Validate missing or malformed JSON fields