[BUGFIX] Reject empty regex_list in ExpectColumnValuesToMatchRegexList

[anxkhn]

ExpectColumnValuesToMatchRegexList declared its regex_list field with no
non-empty validator, so constructing it with an empty list succeeded:

import great_expectations.expectations as gxe

# Accepted today, no error at construction:
gxe.ExpectColumnValuesToMatchRegexList(column="my_col", regex_list=[])

At validate time the pandas metric builds its match frame with
pd.concat(regex_matches, axis=1, ...). With an empty regex_list the loop
that populates regex_matches never runs, so the call becomes
pd.concat([], axis=1) and raises ValueError: No objects to concatenate. This
surfaces to the user as a MetricResolutionError with the opaque message
"No objects to concatenate", which gives no hint that the real problem is the
empty regex_list.

This is an asymmetry rather than a design choice:

• The SQLAlchemy branch of the same metric already guards it
  (column_values_match_regex_list.py: raises "At least one regex must be
  supplied in the regex_list.").
• The mirror expectation ExpectColumnValuesToNotMatchRegexList already has a
  @pydantic.validator("regex_list") that rejects empties with "regex_list must
  not be empty". The like_pattern_list expectations have the same guard.

Only ExpectColumnValuesToMatchRegexList was missing the construction-time
validator, so its empty-list input reached the unguarded pandas pd.concat and
crashed.

Change

• Add a @pydantic.validator("regex_list") to ExpectColumnValuesToMatchRegexList
  that raises ValueError("regex_list must not be empty") when the list is empty.
  It is identical to the validator already used on the not-match sibling, so the
  two expectations now behave consistently and fail fast at construction with a
  clear message. Non-empty lists are unaffected.
• Add a @pytest.mark.unit regression test mirroring the not-match sibling's
  test_invalid_config.

How it behaves after the change

import great_expectations.expectations as gxe

gxe.ExpectColumnValuesToMatchRegexList(column="my_col", regex_list=[])
# pydantic.ValidationError: ... regex_list must not be empty

gxe.ExpectColumnValuesToMatchRegexList(column="my_col", regex_list=["^a"])
# still constructs fine - no behavior change for valid input

Checklist (from the template)

• Description of PR changes above includes a link to an existing GitHub issue

[netlify]

✅ Deploy Preview for niobium-lead-7998 canceled.

Name	Link
🔨 Latest commit	7421b14
🔍 Latest deploy log	https://app.netlify.com/projects/niobium-lead-7998/deploys/6a4e4eb24ffe0100081e6f16