added tests for the new regex (#86)

Added the needed test for alchemlyb PR # 273,
namely:
a field=value without spaces around the equal sign, and a file where the initial simulation time can't be read.

src/alchemtest/amber/access.py

@@ -146,6 +146,8 @@ def load_testfiles():
         "no_control_data": [testfiles_path / "no_control_data.out.bz2"],
         "no_atomic_section": [testfiles_path / "no_atomic_section.out.bz2"],
         "high_and_wrong_number_of_mbar_windows": [testfiles_path / "high_and_wrong_number_of_mbar_windows.out.bz2"],
+        "no_spaces_around_equal": [testfiles_path / "no_spaces_around_equal.out.bz2"],
+        "no_starting_simulation_time": [testfiles_path / "no_starting_simulation_time.out.bz2"],
         }
 
     with open(testfiles_path / 'descr.rst') as rst_file:

src/alchemtest/amber/testfiles/descr.rst

@@ -15,6 +15,8 @@ Notes
 - no_useful_data.out.bz2: AMBER output file without useful data, truncated after the header
 - none_in_mbar.out.bz2: AMBER output file with a wrongly formatted MBAR section. Specifically, a lambda value in a MBAR section has been altered, so it dowsn't match with the other MBAR sections and the expected lambda values (0.2500 --> 0.2550)
 - not_finished_run.out.bz2: AMBER output file from an unterminated run
-- high_number_of_mbar_windows.out.bz2: AMBER output file from  a run with high number of MBAR lambdas
+- high_number_of_mbar_windows.out.bz2: AMBER output file from a run with high number of MBAR lambdas
+- no_spaces_around_equal.out.bz2: AMBER output file where there are no spaces around the '=' # the ' begin time read from' section
+- no_starting_simulation_time.out.bz2: AMBER output file where the starting simulation time is not read
 
 .. versionadded:: 0.7.0