dbconsole: link to "nodes" from statement activity confuses KV nodes and SQL instances under multitenancy

**Describe the problem**

I created a 3-node cluster with 5 SQL instances running in separate processes. I also _restarted_ the SQL instances once to ensure they use fresh instance IDs.

So my node IDs are n1,n2,n3.
The SQL instance IDs are 19,20,21,22,23

When I navigate to DB Console for my SQL activities, then navigate to one of the query fingerprints, I see the query was running on "Nodes: N20,N21,N22".

Two problems:

- It's somewhat unclear what the word "nodes" refers to.
- The hyperlink from the text "N20" in this example links to `/nodes/20`, which is an invalid URL. The `/nodes` URLs refer to KV nodes, not SQL instances, and there is no KV node with ID 20 in this example.

**Expected behavior**

- We should choose which words refer to which "kind" of node.
- The hyperlink from SQL insights should link to a page that displays details about SQL instances, not KV nodes.

**Example in a test cluster**

Jira issue: CRDB-28003

Reproduce the DB Console path from SQL activities to a query fingerprint and inspect the “Nodes” labels and links shown there. Trace the SQL insights link target and confirm that the wording distinguishes KV nodes from SQL instances; done means the link is valid and opens SQL instance details.

Written by the indexing model from the issue text.