Modify code to handle non-standard and time-varying component orientations

[ShangqinHao]

Thank you for providing this package, which has significantly improved the efficiency of my PRF processing.

I made two changes in this pull request:

1. Correction to SAC Header cmpinc Calculation
   The relationship between the dip defined in the SEED standard and the SAC header cmpinc should be cmpinc = dip + 90 rather than the previously assumed cmpinc = −dip. While this error does not impact data using standard NEZ components, it becomes significant when rotating non-standard components to NEZ, where correct "dip" values are required. To address this, I modified the rotateZNE function in eq.py and the _add_header function in rf.py.

2. Handling Time-Varying Component Orientations
   The original _add_header function could not correctly process stations with component orientations that vary over time (e.g., station II.MSVF). I updated this function to determine the appropriate orientation based on the correct time period for each station.

I tested the updated code using station II.MSVF, which has non-standard and time-varying component orientations:

ch.start_date, ch.end_date, ch.azimuth, ch.dip:
1994-05-24T00:00:00.000000Z 2002-06-05T23:59:59.000000Z 180.0 0.0
2002-06-06T00:00:00.000000Z 2007-06-15T23:59:59.000000Z 180.0 0.0
2007-06-16T00:00:00.000000Z 2008-05-12T23:59:59.000000Z 180.0 0.0
2008-05-13T00:00:00.000000Z 2010-07-21T23:59:59.000000Z 180.0 0.0
2010-07-22T00:00:00.000000Z 2013-01-17T23:59:59.000000Z 180.0 0.0
2013-01-18T00:00:00.000000Z 2013-10-05T05:24:59.000000Z 169.8 0.0
2013-10-05T05:25:00.000000Z 2016-10-16T23:59:59.000000Z 169.8 0.0
2016-10-17T00:00:00.000000Z 2018-03-06T19:59:59.000000Z 169.8 0.0
2018-03-06T20:00:00.000000Z 2022-06-16T03:29:59.000000Z 169.8 0.0
2022-06-16T03:30:00.000000Z 2022-08-06T23:59:59.000000Z 169.8 0.0
2022-08-07T00:00:00.000000Z None 67.1 0.0

Thank you for reviewing this pull request, and I look forward to your feedback!

[xumi1993]

Hi Shangqin,
Great to hear from you. Thanks for your improvement on Seispy. These changes are very important for the efficiency of RF calculations. But I have changed the branch to dev. I'll merge all changes from dev to master at the next version.

[ShangqinHao]

Got it. I will submit a pull request to your dev branch the next time I modify the code.