bug(health): Redfish SSE applies to Switch Hosts

[mkoci]

Version

0.8.3

Describe the bug.

When using the Server-Sent event subscription for BmcEndpoints, no distinction is made between Bmc Redfish servers and Switch hosts. We should not run BMC specific discovery and collection on Switches.

Minimum reproducible example

Enable 

[collectors.logs]
mode = "sse"

In the config for health when you also have Switch hosts defined in static configuration or ingested and available for discovery through NICo.

### Relevant log output

```shell
level=ERROR msg="Failed to discover entities" error="BmcError(BMC error: BMC Error: Invalid HTTP response - url: https://10.85.14.106/redfish/v1 status: 4
level=ERROR msg="Error during collector iteration" collector_type= endpoint="BmcAddr { ip: 10.85.14.106, port: Some(443), mac: MacAddress

Other/Misc.

No response

Code of Conduct

• I agree to follow NCX Infra Controller's Code of Conduct
• I have searched the open bugs and have found no duplicates for this bug report

[yoks]

Not sure what i understand what this bug is. NICo should work witch switches BMC, this is how it disovers swtich sensors and leak sensors data. If it is specific for SSE, it should not have distinct logic to spawn switch vs non swtich collectors. There is arleady metadata available for this.