Fix handling of excerpt regions in EP CLI (#49936)

Previously, we were not computing excerpt regions correctly for EP
examples captured from prod. This PR fixes that, and also simplifies the
data flow in the EP CLI. Examples either come from a concise spec (like
the markdown evals), or are collected from prod. Either way, we compute
from them a `ZetaPromptInput`, and the downstream steps like
prompt-formatting and scoring are derived from that.

Release Notes:

- N/A

---------

Co-authored-by: Ben Kunkle <ben@zed.dev>

crates/edit_prediction_cli/src/split_commit.rs

@@ -371,7 +371,7 @@ pub fn generate_evaluation_example_from_ordered_commit(
         reasoning: None,
         uncommitted_diff: String::new(),
         rejected_patch: None,
-        captured_prompt_input: None,
+
         telemetry: None,
         human_feedback: Vec::new(),
         rating: None,
@@ -1370,7 +1370,7 @@ Date: Mon Jan 1 00:00:00 2024
             reasoning: None,
             uncommitted_diff: String::new(),
             rejected_patch: None,
-            captured_prompt_input: None,
+
             telemetry: None,
             human_feedback: Vec::new(),
             rating: None,