[Improvement]: Improve AmsThriftUrl parsing and validation

### Search before asking

- [x] I have searched in the [issues](https://github.com/apache/amoro/issues?q=is%3Aissue) and found no similar issues.

### What would you like to be improved?

### Motivation

`AmsThriftUrl` is used by clients to parse AMS thrift and ZooKeeper URLs. The current implementation has a few issues that can make URL parsing less robust:

1. `parserThriftUrl()` lowercases the entire URL before parsing.
2. `socketTimeout` query parameter parsing is duplicated.
3. thrift URL validation is weak for missing host, missing port, or unsupported scheme.
4. Some error messages are too generic, such as `invalid ams url`.

Lowercasing the whole URL may unexpectedly change case-sensitive parts, such as catalog names or query parameter values.

### Current behavior

For example:

```java
AmsThriftUrl.parse("ThRiFt://LOCALHOST:1260/MyCatalog?socketTimeout=6000", null)
```

currently lowercases the URL before parsing, so values such as host/catalog may be changed unexpectedly.

Invalid URLs such as the following are not clearly rejected during parsing:

```
http://127.0.0.1:1260/catalog
thrift:///catalog
thrift://127.0.0.1/catalog
```

### Expected behavior
AmsThriftUrl should:

- Preserve the original URL case for host, catalog name, and query values.
- Treat URL schemes case-insensitively.
- Validate that thrift URLs use the thrift scheme.
- Validate that thrift URLs contain both host and port.
- Parse `socketTimeout` consistently in thrift and ZooKeeper URL paths.
- Provide clearer error messages for invalid URLs.

### Proposed changes
- Avoid calling `toLowerCase()` on the whole URL.
- Parse URL scheme with case-insensitive comparison.
- Add explicit thrift URL validation for:
- unsupported scheme
- missing host
- missing port
- Extract shared socketTimeout query parsing logic.
- Add unit tests for AmsThriftUrl.

### How should we improve?

_No response_

### Are you willing to submit PR?

- [x] Yes I am willing to submit a PR!

### Subtasks

_No response_

### Code of Conduct

- [x] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)

Start by locating the AmsThriftUrl class and its parse, parserThriftUrl, and ZooKeeper URL paths. Review the existing URL parsing and validation, then add unit tests covering case preservation, case-insensitive schemes, required host and port, supported schemes, consistent socketTimeout parsing, and clearer errors; done means these cases are validated by the tests.

Written by the indexing model from the issue text.